<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Ortaya Karışık (Fatma Betul Dincaslan) on Medium]]></title>
        <description><![CDATA[Stories by Ortaya Karışık (Fatma Betul Dincaslan) on Medium]]></description>
        <link>https://medium.com/@fatmab-dincaslan?source=rss-d035227fb6ad------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*M5Xfd2g5RlZ5Xkjd6tDsww.png</url>
            <title>Stories by Ortaya Karışık (Fatma Betul Dincaslan) on Medium</title>
            <link>https://medium.com/@fatmab-dincaslan?source=rss-d035227fb6ad------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 16:51:09 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@fatmab-dincaslan/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[FoG2026London: From a Perspective]]></title>
            <link>https://fatmab-dincaslan.medium.com/fog2026london-from-a-perspective-bfcf19fb9d6c?source=rss-d035227fb6ad------2</link>
            <guid isPermaLink="false">https://medium.com/p/bfcf19fb9d6c</guid>
            <dc:creator><![CDATA[Ortaya Karışık (Fatma Betul Dincaslan)]]></dc:creator>
            <pubDate>Tue, 14 Jul 2026 06:39:58 GMT</pubDate>
            <atom:updated>2026-07-14T06:39:58.093Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>If you remember, I joined FoGLondon for the first time last year. In this post, I will share my notes and summaries from the talks I attended or discovered while roaming at the conference this year.</blockquote><h3>What Was It About (#FoG2026London)</h3><p>You can find the detailed description in my previous post <a href="https://medium.com/@fatmab-dincaslan/fog2025london-from-a-perspective-1e18f7f8bd26">here</a>.</p><p>It covers many aspects of drug discovery/development, genome editing, rare diseases, single-cell, microbes, sequencing, and transcript/epigen/gen/prote/metabol/metagen/pharmacogen/spatial/multi-omics, generating/storing/managing/using/distributing the big data.</p><p>This year, there was a stage for Advanced Therapies as well. You can find more details on the event webpage (<a href="https://festivalofgenomics.com/london">https://festivalofgenomics.com/london</a>).</p><p>I used to be a single-cell scientist during my PhD. Everything has changed rapidly in the last few years. I wanted to keep up with the field as well as follow the trends in spatialomics. If you know me long enough, you might be aware of my review on spatialomics &gt;5 years ago, before it got popular (here is the link for my detailed review, <a href="https://medium.com/@fatmab-dincaslan/spatialomics-past-today-and-future-1490a43d8a4d">https://medium.com/@fatmab-dincaslan/spatialomics-past-today-and-future-1490a43d8a4d</a>). I was wondering what has been changing in the field. Moreover, I have a personal and professional interest in therapeutics.</p><p>I enjoy research and research-focused talks, so I will share mostly my research/technical-oriented notes here.</p><h3>Day 1</h3><p>I listened to the talks in the main, single-cell, spatial, advanced therapies, genomic medicine, proteogenomics, and pharmacogenomics stages during the day.</p><ul><li>Spatial Profiling at Scale: Where Are We Now and What Comes Next? (by Jasmine Plummer)</li></ul><p>The talk covered the journey of transcriptomics: bulk -&gt; single-cell-&gt; spatial. Dr. Plummer described the main spatial technologies under three sub-categories:</p><ol><li>Sequencing-based barcoded surfaces</li><li>Imaging-based/in-situ sequencing with combinatorial probes</li><li>Multiplexed antibody (ab) imaging-based/barcoded antibodies</li></ol><p>A combination of multi-omics and several omics technologies might provide newer insights (though I think it might be expensive to apply). Moreover, spatialomics enables the detection of rare cell types in a single slide. She mentioned Gene Regulatory Networks (GRN) in the context of neurodevelopmental disorders. For example, the spatial specificity of transcription factors (TF) for the localisation and identification of cancer origins/niches.</p><ul><li>Complementary, scalable single-cell and spatial profiling to reveal deeper biological insights (by Sarah Taylor)</li></ul><p>If you are familiar with 10X technologies, you might be aware of the Chromium controller for single-cell encapsulation, as well as spatial technologies. Visium and Xenium are two major spatial platforms that 10X has. Basically, Visium uses probe hybridisation to spatially barcoded slide, later allowing high-throughput sequencing, whereas Xenium is a sort of fluorescent in situ hybridisation (FISH)-based technique depending on imaging/image resolution of high-plex targeted panels (suggested reading, <a href="https://www.omicsempower.com/blog/10x-genomics-visium-hd-vs-xenium/#:~:text=%2C%20Right:%20Xenium)-,01.,detected%20usingthree%20probes%20per%20gene">https://www.omicsempower.com/blog/10x-genomics-visium-hd-vs-xenium/#:~:text=%2C%20Right:%20Xenium)-,01.,detected%20usingthree%20probes%20per%20gene</a>.).</p><p>Sarah mentioned the advantages and applications of the technologies above and GEM-X Flex v2 in the context of the tumor microenvironment (TME). She also explained the combination of GEM X-Flex (<a href="https://www.10xgenomics.com/products/flex">https://www.10xgenomics.com/products/flex</a>) and Xeinum barcoded panels for more powerful analysis.</p><ul><li>Pharmacogenomics and Healthcare (by Sir Mark Caulfield)</li></ul><p>Dr. Caulfield discussed the importance of inclusion and considering diversity in the context of pharmacogenomics and drug discovery in the 21st century.</p><ul><li>Getting Commissioned ATMPs* to Patients: Key Insights (by Ben Doak)</li></ul><p>The well-known therapeutic options were listed as</p><ol><li>Gene therapy</li><li>Enzyme replacement</li><li>Bone marrow transplantation</li><li>Small tags for therapeutics localisation</li></ol><p>The talks covered the newborn (metabolic) diseases and the importance of early diagnosis and treatment. I think it was amazing to see how efficient these treatments could be if applied timely and wisely (I mean, based on focusing on a specific pathway the given disease is associated with). The examples were given/applied to fatty acid metabolism and lysosomal diseases.</p><p>For the effective therapies, the speaker highlighted two parameters:</p><ol><li>Where in the pathway does the disease originate (this requires careful consideration for each patient)</li><li>The level of deficiency</li></ol><p>*ATMP: Advanced Therapy Medicinal Product</p><ul><li>Raising the Bar in Rare Disease: Networks of Excellence, New Technologies and New Standards (by Emma Baple)</li></ul><p>The NHS’s approach to rare diseases to accelerate and increase the precision and efficiency of the process using multiomics, epigenomics, and AI technologies was discussed.</p><ul><li>The Evolving Sequencing Landscape: How All the New Technologies Stack Up (by Michel Quail)</li></ul><p>I think this was the talk I wanted to attend from the beginning. Dr. Quail briefly introduced each technology and platform and then compared them with the pros and cons of each of them. These include short and long read sequencing, direct RNA sequencing, and multiomics approaches.</p><p>In addition to well-known platforms for short and long-read sequencing (e.g., Illumina, Oxford Nanopore, Pac-Bio), he explained the newer platforms (e.g., Roche Axelorois-IN (<a href="https://sequencing.roche.com/global/en/article-listing/sbx-platform-reaches-unprecedented-speeds.html">https://sequencing.roche.com/global/en/article-listing/sbx-platform-reaches-unprecedented-speeds.html</a>), Ultima (<a href="https://www.ultimagenomics.com/">https://www.ultimagenomics.com</a>), Element Aviti24 (<a href="https://www.elementbiosciences.com/products/aviti24">https://www.elementbiosciences.com/products/aviti24</a>)).</p><p>For example, Roche uses a nanopore-like approach in terms of pores and currents while sequencing them by expansion (SBX, using Xpandomers with cleavable bonds, allowing translocation control, base-specific reporting, and enhancements of signal). The main advantage seems to be speed and 80% full gene coverage for long reads. On the other hand, Ultima is a single-end technology providing low-cost, flow-cell-based sequencing. Their technology is called mostly natural sequencing by synthesis (mnSBS) due to optical scanning and current levels for bases, and dependency on a few fluorescently labeled bases. Element Biosciences also entered the market with AViti24, which relies on direct in-sample sequencing (DISS, in situ), Avidity Base Chemistry (ABC), and easy multiomics integration of RNA, DNA, and protein detection.</p><ol><li>Roche for RNA/full-length (flexible and reusable chip)</li><li>Ultima, for circulating tumor DNA (ctDNA) with PPM-Seq (<a href="https://www.ultimagenomics.com/products/ppmseq-tm/">https://www.ultimagenomics.com/products/ppmseq-tm/</a>)</li><li>Illumina constellation for somatic variant detection</li><li>Element for high-plex and low-cost experimental design.</li></ol><p>He briefly mentioned Thermo Fisher Ion Torrent as an alternative for short-read sequencing and clinical variant analysis.</p><ul><li>From immune repertoires to designed antibodies: AI models for nativeness, humanisation, and therapeutic design (by Pietro Sormanni)</li></ul><p>Dr. Sormanni introduced AIRR-seq for antibody-omics and the challenges of antibody (ab) development that necessicates appraoches beyond natural ones under two main categories:</p><ol><li>Functionality (e.g., binding to epitope, cross-reactivity, activity at lower doses)</li><li>Developability (e.g., shelf-life/stabil-storage and serum-life)</li></ol><p>Although it might be easier to develop in vitro, in vivo production of abs were preferred due to immune properties (e.g., low immunogenicity).</p><p>Their approach, AbNatiV/AbNatiV2, <a href="https://gitlab.developers.cam.ac.uk/ch/sormanni/abnativ">https://gitlab.developers.cam.ac.uk/ch/sormanni/abnativ</a>, assesses such properties (e.g., humanness and nativeness) to maximize the compatibility of in-vitro abs using the sampled mutation screen for nanobody humanisation.</p><ul><li>The Power to Predict and to Perceive: AI-Based Discovery in Proteomics Data (by Lennart Martens)</li></ul><p>Dr. Martens explained and exemplified AI-based discovery (ML and GNNs) for proteogenomics data. He highlighted the importance of modifications as well as sequences for the contribution of diversity. In this context, he shared the tools their team developed to predict the behavior of molecules.</p><ol><li>MS2PIP to predict peptide fragmentation, <a href="https://iomics.ugent.be/ms2pip/">https://iomics.ugent.be/ms2pip/</a></li><li>DeepLC to predict the retention time of proteins and on diverse modifications, <a href="https://www.nature.com/articles/s41592-021-01301-5">https://www.nature.com/articles/s41592-021-01301-5</a></li><li>IM2Deep to predict ion mobility of modified peptides, <a href="https://github.com/compomics/IM2Deep">https://github.com/compomics/IM2Deep</a></li><li>MS2Rescore to improve the sensitivity in the complex settings, <a href="https://ms2rescore.readthedocs.io/en/stable/">https://ms2rescore.readthedocs.io/en/stable/</a> (particularly useful for single-cell proteomics)</li><li>lessSDRF (Sample and Data Relationship Format), <a href="https://pmc.ncbi.nlm.nih.gov/articles/PMC10598006/">https://pmc.ncbi.nlm.nih.gov/articles/PMC10598006/</a>, to improve annotations</li><li>PRIDE (Proteomics Identifications) database, and Scop3PTM, a database for post-translational modifications (PTMs)</li><li>PTMVision for single-sample PTM discovery, <a href="https://pubs.acs.org/doi/full/10.1021/acs.jproteome.4c00679">https://pubs.acs.org/doi/full/10.1021/acs.jproteome.4c00679</a></li><li>MLMarker to derive which tissue it is derived from, <a href="https://www.biorxiv.org/content/10.1101/2025.06.10.658642v1">https://www.biorxiv.org/content/10.1101/2025.06.10.658642v1</a></li><li>MoDPA for modification-dependent encoder for protein association, <a href="https://www.biorxiv.org/content/10.64898/2026.01.20.700550v1">https://www.biorxiv.org/content/10.64898/2026.01.20.700550v1</a></li><li>Tabloid proteome for the visualization and exploration of MoDPA results altogether and alternative products of the so-called dark proteome, such as small open reading frames (sORFs), <a href="https://iomics.ugent.be/tabloidproteome/">https://iomics.ugent.be/tabloidproteome/</a></li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*aCnUmNGXpRX5buZ3" /></figure><p>The best freebie shirt I have ever gotten in my life :) (credit: <a href="https://plasmidsaurus.com)/">https://plasmidsaurus.com)</a></p><h3>Day 2</h3><p>I listened to the talks in single-cell, spatial, multi-omics, and AI in healthcare stages during the day.</p><ul><li>GBM-Space: Dissecting the Cellular Circuits of Glioblastoma Using Single Cell and Spatial Genomics (by Omer Bayraktar)</li></ul><p>Dr Bayraktar described the GBM-Space,<a href="https://www.gbmspace.org/">https://www.gbmspace.org</a>, under 5 main categories:</p><ol><li>Spatial tumor organization</li><li>Cell trajectories</li><li>Clonal organization</li><li>Cell intrinsic TME signals</li><li>Gene regulation and plasticity</li></ol><p>You start with single-cell (sc) data, continue with malignant cell states with scRNA-seq and atac-seq, then follow the path for tumor organization using Visium spatial data. Whole Genome Sequencing (WGS) might be useful for Single-nucleaotide variant (SNV) associations to confirm clonal trajectories from RNA-seq data. Later with the patient derived organoids can be used to recapitulate cell states and organization. Overall, this might allow you to discover tumor tissue niches with cell trajectories.</p><ul><li>Ushering in a New Era of Biological Discovery With Scaled Up Single Cell Sequencing (by Mike Day)</li></ul><p>I might have fallen in love with the passion of Mike :) He shared their story of making Parse from zero to kinda hero. If you are not aware, the company is now acquired by Qiagen.</p><p>He also introduced his new product: Giga-Tahoe, <a href="https://www.parsebiosciences.com/news/tahoe-therapeutics-selects-parse-biosciences-gigalab-to-generate-300-million-single-cell-profiles-for-large-scale-perturbation-atlas/">https://www.parsebiosciences.com/news/tahoe-therapeutics-selects-parse-biosciences-gigalab-to-generate-300-million-single-cell-profiles-for-large-scale-perturbation-atlas/</a>. He mentioned that sequencing of millions of cells with relatively cheaper price might allow homogeneous and more reproducible results, <a href="https://www.biorxiv.org/content/10.1101/2025.02.20.639398v1.full">https://www.biorxiv.org/content/10.1101/2025.02.20.639398v1.full</a>.</p><ul><li>Single-Cell Phenotyping and DNA Profiling to Deconvolute Forensic Biological Mixtures Originating From Multiple Contributors (by Iain Macaulay)</li></ul><p>This was an interesting possible application of single cell approaches.</p><p>Dr. Macaulay mentioned that the forensic evidences are usually mixture of cell and DNA mix. He showed that the single-cell phenotyping with DNA sequencing is possible by distingusihing subtle differences. They succesffuly deconvulated 6-single-cell mix from 6-different individuals. He also revealed that rather than whole transcriptome, mitocondrial RNA (mtRNA) of single-cells might be more informative for forensic evidence.</p><p>I am wondering what happens, if they do this for complete RNA-seq, and include all rRNA, tRNA, snRNA and so on…</p><ul><li>Agentic AI Reasoning in Biomedicine (by André Freitas)</li></ul><p>Dr. Freitas talked about AI integration from pharmagenomic reasoning to critical decision-making. One starts with local&amp;integral evidence, then focuses on biological understanding (rather than automating the task).</p><p>He elaborated the difference between general purpose AI agent such as chatGPT or Gemini and domain-knowledge requiring specialised AI agents. He introduced Lunar to easily build your own AI agent, <a href="https://lunarbase.ai/">https://lunarbase.ai</a>.</p><ul><li>SPACE: A Centre Of Excellence To democratize &amp; Scale Up Spatial omics (by Dario Bressan)</li></ul><p>Dr. Bressan started with the spatial technologies under five main categories:</p><ol><li>Sequencing based transcriptomics</li><li>Cyclic imaging based transcriptomics</li><li>Light induced tagging</li><li>Mass-spec based</li><li>Mass cytometry imaging</li><li>Cycling imaging proteomics</li></ol><p>He clamied that the diversity of the field makes it challenging to adopt and use it. That’s why they came up with the idea of SPACE: Spatial Profiling and Annotation Centre of Excellence (including data centre) at Cambridge, <a href="https://www.cancergrandchallenges.org/sites/default/files/2025-03/SPACE_showcase_b.pdf">https://www.cancergrandchallenges.org/sites/default/files/2025-03/SPACE_showcase_b.pdf</a>.</p><p>I think such core facilities are crucial for making science smoother. I am big fan of such organizations.</p><p>He also invited everyone to the upcoming symposium: 1st Grand Challenge Conference in Spatial Biology in September, <a href="https://www.cancergrandchallenges.org/conferences">https://www.cancergrandchallenges.org/conferences</a>.</p><ul><li>Single-Cell and AI/ML to Understand Human Development and Disease (by Muzlifah Haniffa)</li></ul><p>Dr. Haniffa provided insights from the combination such trio-data in omics for therapeutics:</p><ol><li>scRNA-seq</li><li>Xenium</li><li>Histology</li></ol><p>She also mentioned very exciting news for scientific environment in the UK: Welcome Trust 2028 plans for expanding the scientist crew! (more info and job preparedness for what’s coming, <a href="https://www.wellcomegenomecampus.com/thecampus/campusexpansion/">https://www.wellcomegenomecampus.com/thecampus/campusexpansion/</a>). I loved the idea of whole campus, everything included. I think this will bring amazing opportunities for job-seekers.</p><ul><li>Panel Discussion: Bridging the Gap: Integrating Single-Cell and Spatial Transcriptomics for Deeper Biological Insights (panel members were different than stated online)</li></ul><p>The main takeaway message is that one size fits all approach do not work for omics. You need ask some experts before planning, you need to consider which type pf research questions you are interested in and what’s your budget.</p><p>They also discussed mainly the spatial omics, data integration challenges and why it is not ready for clinical settings.</p><p>Because of sequencing costs depending on cell number and depth, we could not even reach to the broader adoption of sequencing technologies. I think spatalomics a big luxury for most researchers.</p><p>What was missing at the conference could be the complete transcriptomics. If we keep focusing on mRNA based transcriptomics and use “non-coding RNA” as auxiliary or still junk, I dont think we are making a progress in the transcriptomics.</p><p>Perhaps we need more research adopting complete/whole/total RNA transcriptome to decipher (RNA) regulatory networks in different contexts.</p><p>Further reading about complete high-throughput transcriptomics and data analysis:</p><ol><li>sc-Complete-seq <a href="https://academic.oup.com/nar/article/53/14/gkaf699/8208114">https://academic.oup.com/nar/article/53/14/gkaf699/8208114</a> (the paper relevant to the poster above)</li><li>Smart-seq total <a href="https://www.pnas.org/doi/10.1073/pnas.2113568118">https://www.pnas.org/doi/10.1073/pnas.2113568118</a></li><li>Vasa-seq <a href="https://www.nature.com/articles/s41587-022-01361-8">https://www.nature.com/articles/s41587-022-01361-8</a></li><li>Beyond Poly-A <a href="https://www.biorxiv.org/content/10.1101/2025.08.08.669394v1.full">https://www.biorxiv.org/content/10.1101/2025.08.08.669394v1.full</a>.</li></ol><p>Overall, I enjoyed and learned a lot (one more time). I want to thank everyone for making this event possible (particularly the Lumb family and the FoG team). #FoG2026 #FoG2026London #FestivalOfGenomics</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*XNh-63EYIXAWpuUS" /><figcaption>Natural History Museum South Kensington (iPhone 12mini)</figcaption></figure><p>I am looking forward to joining the next exciting conference!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bfcf19fb9d6c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Cost of Living for Post-Docs]]></title>
            <link>https://fatmab-dincaslan.medium.com/cost-of-living-for-post-docs-f9f06bd37835?source=rss-d035227fb6ad------2</link>
            <guid isPermaLink="false">https://medium.com/p/f9f06bd37835</guid>
            <dc:creator><![CDATA[Ortaya Karışık (Fatma Betul Dincaslan)]]></dc:creator>
            <pubDate>Mon, 13 Jul 2026 00:38:04 GMT</pubDate>
            <atom:updated>2026-07-13T00:38:04.073Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>I have been planning to make some comparative charts for the cost of living in different locations since the beginning of the postdoc. If you have agentic AI, you know you can do more than that, so why not? Here you will see the results with me. DISCLAIMER: I am not a financial expert. If you notice anything significantly wrong or misleading, please let me know via DM. Also, please feel free to update these with more reliable resources if you have any.</blockquote><p>In postdocs, the financial situation is much better compared to PhD for sure. However, how much better is a big question mark. We can only say that it is better compared to PhD. If you think about the relocation cost, constant instability, and cost of living, you will have a better picture.</p><p>Considering age will also be an important component while looking for jobs or a pension scheme, everything eventually matters.</p><p>Let’s clarify: it is not whining about academia. There is no harm in dreaming big while also being realistic and strategic. If there is something that can be improved, why not give it a shot!</p><p>As someone intrigued by data science and charts, I quickly prompted agentic AI tools to create these beautiful charts. I kinda cross-checked based on my experiences. Yet, you are always welcome to make it better and more reliable.</p><h3>Postdoc salaries are variable</h3><p>Some places treat post-docs as staff and working class, so it is subject to tax. Depending on how much you earn, the tax amount might vary from one country to another.</p><p>However, in some countries, the position is considered an extension of PhD studentship, so you are paid a bursary (not salary), usually less than a staff salary, and you have fewer benefits.</p><p>While some countries simply focus on the innovation, due to the post-doc’s significance in moving bio/tech forward, even if there is tax, it might be negligible.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*BpHWWPfZ7jyLHcee" /><figcaption>Gross vs net salary (after tax) based on 34 cities across 22 countries. You can see the estimated tax rate for each country/city as well. Please feel free to double-check.</figcaption></figure><p>Overall, net salary-wise, Switzerland is the winner. It is followed by Singapore and the United Arab Emirates.</p><p>However, the cost of living differs from one city to another. Next, we might ask whether you can live a decent life with these salaries in the respective cities.</p><p>To be fair, and having similar standards, I considered studio rent price for a single person for accommodation (although it is still a luxury in most places), the remaining income after you spend for the basics, and the healthcare standards.</p><h3>Which cities can you have a decent life with it?</h3><p>The estimated living cost is a little bit subjective here. If you include childcare, etc., it might differ. So be careful while interpreting it.</p><p>Let’s start with the living wage. Although salaries above in some countries might be tempting, we need to make sure whether it is above or below the living wage.</p><p>Not-so-fun fact: most of the time, PhD salaries/bursaries usually can’t make it to the living wage.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/929/0*5CwWuU7FiSCytpnh" /><figcaption>Living wage vs. postdoc salaries in the selected cities.</figcaption></figure><p>Well, good news, post-doc salaries are usually above the living wage or at least close to it in some others! Germany, Portugal and Turkey might be still in the game with Switzerland, Singapore, China and UAE. Unfortuately, in some cities in the US, such as SF, or even the UK, the disparity seems concerning. Is this true?</p><p>Let’s continue with housing affordability with a standard studio rent price.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*5r8kMYvVZCJjveQk" /><figcaption>30% of salary paid for the rent is used as the standard housing affordability metric. If it is above 30%, it is considered as financially stressful. Please keep in mind that there might be slight deviation in the up-to-date mettrics.</figcaption></figure><p>It seems, you cannot afford living in the studio in some cities such as San Francisco, London or Boston with a nationwide postdoc salary. Similarly, it is hard to live alone in majority of these locations.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*LS3tdoYIHnzr3lot" /><figcaption>Disposable income is the green bars showing whatever remains that you can save for other things in life after rent, groceries, transport and utilities. The longer the green portion, the wealthier you might feel in that city.</figcaption></figure><p>Although, the salaries/bursaries are low in some cities like Delhi or Shanghai, it seems you can still make it with nationwide post-doctoral salary. I had no idea about this before, so it was suprising for me as well. (However, please keep in mind that, the international conferences etc., might be still costly when you convert the currencies).</p><p>Let’s make a fancier representation of these.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*2cRYMqmRJ8LFzIgZ" /><figcaption>The bigger and higher on the y-axis the bubble, the bigger the leftover from the postdoc salary. Switzerland is rocking in many ways.</figcaption></figure><p>The bubble charts helps us to visualize the economically advantageous position of doing post-doc in certain cities and countries. Hence, despite the high living cost (the x-axis) in certain countries such as Switzerland or Singapore, the post-doctoral position is still economically advantageous.</p><p>Well, that sounds interesting. However, we know that there are more opportunities in some countries beyond government/national postdoc. Let’s dive into them!</p><h3>Alternative postdoctoral options</h3><p>As you might aware that, the post-doc position is not limited to academia. So, you might have industry post-doc options in some countries with slighlty better salary plus industry experience! A better academic option can be international fellowships such as Marie Curie (MSCA).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/924/0*5zXFkQPdTHadKAtt" /><figcaption>A representative graph for alternative post-doctoral positions and respective salaries in the selected cities. This graph can be updated with other less-known international fellowships.</figcaption></figure><h3>How about the healthcare?</h3><p>Usually post-docs are above 28, and mostly above 30 years old. At this age you start to think about a decent healthcare in a country you are living in.</p><p>Although, inter/national post-doc positions often ignore a comprehensive health insurance, in some countries you might still get a reasonable care with affordable prices. The leftover from your salary might be useful here.</p><p>I evaluated the healthcare system under 4 categories: quality, accessibility, affordability and dental affordability (it is seperate in most countries).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/924/0*uGvr2FxIcUDhsn5U" /><figcaption>WHO and Numbeo-based metrics to assess the healthcare system. The higher the score, the better the system. Please feel free to update the graphs if you have better resources.</figcaption></figure><p>Even if you do not earn higher salaries with respect to minimum wage workers in Tokyo, it seems you are in good hands for the healthcare system.</p><p>Now, I am not suprised why Switzerland and Singapore are one of the leading countries to attract the tip of the global talent.</p><h3>How stable is your income?</h3><p>Post-doc salaries usually do not change within the year. Therefore, it is important that whatever you are earning is not melting simply because of the inflation.</p><p>Although, the salary and healthcare system might be tempting, if the economy is declining, the stressfull economical post-doc life might turn into a survival game. A real one!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/924/0*K_9txBPvD9n4UcCE" /><figcaption>Average consumer-based inflation. The higher the inflation risk, the less power of purchasing and value of the earning. Source: 2020–2025 WB/IMF.</figcaption></figure><p>I am not sure about how much reliable the data above is but it seems the inflation might be a significant problem for yearly payments.</p><p>If your salary is constant during the year but the inflation is keep rising, whatever you are earning becomes less and less. It might be major challenge in some countries such Turkiye, Egypt or Poland listed above.</p><p>The lower inflation risk in South Korea, Switzerland or Japan might be good examples of stable economies.</p><p>It is important to be able to focus on research rather than thinking how to make ends meet.</p><h3>Final Remarks</h3><p>The majority of post-docs enjoy doing science itself.</p><p>Not only experts in certain fields but also explorers, maybe more like hitchhikers.</p><p>They grab their backpack, and travel one lab to another. And start discovering again.</p><p>We know the science is mostly about the ecosystem, infrastructure and fundings.</p><p>Yet, it might also be more stability about the future.</p><p>Only 1–3 of every 10 post-docs will possibly land in academic position (depending on the country and the funding opportunities for the field). After 5–10 years of post-doc without prior and continous experience, the industry might be a less feasible option. The alternatives will not be easy other. So, the stress is real. The financial instability does not make it easier.</p><p>Hosuing crisis is a huge challenge in many countries. Even providing relocation and housing support could be a good start.</p><p>To thrive in biotech and innovation, perhaps it is time for some countries to invest more to the people who make the change. You want to attract the global talent? Please feel free to invest in scientists as well as science!</p><h3>Reference Tables</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*sVD3AawrDWnp7DnX" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*t_FqBiOmzE_r35GD" /></figure><p>HC: healthcare. col index: numero cost of living.</p><h3>Data Sources</h3><ul><li>Salaries: NIH NRSA FY2025 ($62,652/yr = $5,221/mo, NOT-OD-25–105); Research Ireland 2025–2026 New PostDoc P1 (€47,273/yr); SNSF Postdoc.Mobility (~CHF 7,800/mo); DFG TVL-E13 Stufe 1 (~€4,759/mo, 2025/26); UKRI 2025 Grade 6 min (~£40,497/yr); FCT 2025 junior researcher (~€2,950/mo); FWF; Academy of Finland; VR Sweden (~SEK 42,000/mo); JSPS PD ¥362k/mo; NRF Korea ~KRW 3M/mo; CNRS; A*STAR/NRF Singapore ~SGD 5,900/mo; NSERC PDF CAD 70,000/yr; IIT Delhi/DST-SERB (~INR 82k+HRA); CONAHCYT ~MXN 32k/mo; NRF South Africa ~ZAR 35k/mo</li><li>MSCA-PF (2023–2025 WP, 2025 call): living allowance €5,990/mo × country correction coefficient + €710 mobility. CCCs: Switzerland 163.8%, UK 141.7%, Norway ~145%, Ireland 110%, Sweden 112%, Germany 101.2%, Spain 95.6%, Portugal 86%, Greece 80%, Poland 77%, Turkey 68%. Switzerland MSCA-eligible from Jan 2025 (transitional arrangement; SERI backstops funding if formal association agreement not signed by grant date). Amounts converted at Feb 2026 EUR/USD = 1.18.</li><li>Rents: Numbeo 2025 studio apartment (not 1-bedroom) near university. Cross-checked: <a href="http://daft.ie/">Daft.ie</a> (Dublin), Rightmove (UK), Immoscout24 (DE), Idealista (ES/PT), Magicbricks (Delhi). European costs reflect EUR/GBP/CHF strengthening vs USD (~7–15%) since 2023–2024.</li><li>Turkey: Horizon Europe associated country (TUBITAK NCP). MSCA-PF CCC ~68% → €5,990×0.68+€710=€4,783/mo≈$5,644 at EUR/USD 1.18. TUBITAK 2218 fellowship TRY 43,500/mo (~$1,175 at ₺37/USD, 2025). TUBITAK 2218 stipend is tax-exempt under Turkish law (fellowship income is not classified as employment wage; no income tax deducted). Employed postdocs join SGK from day 1 (work permit), providing universal healthcare incl. basic dental; SGK covers ~99% of population.</li><li>Tax status by program (verified 2025): Tax-exempt: TUBITAK 2218 (Turkey — fellowship not classified as wage income); JSPS overseas fellowship (Japan — explicitly “not employment-based wages, exempted from taxation” per JSPS guidelines). Taxable: NIH NRSA (US — “subject to federal and state income tax,” reported on Form 1099, per NIAID); NSERC PDF / Banting (Canada — explicitly taxable since 2010 CRA ruling, Box 105 T4A); UKRI, DFG TVL-E13, Research Ireland (employment contracts, fully taxable as salary); SNSF Postdoc.Mobility (Swiss income tax applies); MSCA-PF (taxed as employment income in host country, at host-country rates). Japan domestic JSPS PD: classified as 雑所得 (miscellaneous income), not salary — no source withholding, no social insurance contributions; effective rate ~13% after ¥480k basic deduction, lower than the ~24% applied to salaried institutional postdocs.</li><li>Dental affordability: scores reflect cost of routine dental care relative to postdoc salary. Sources: NHS dental charges (UK), SGK Turkey, NHIS Korea/Japan, Numbeo dental procedures, Expatistan dental index 2025. Low-cost leaders: Turkey, India, Mexico (medical tourism destinations). High OOP: US, Switzerland, Ireland (no adult public dental plan).</li><li>Inflation: World Bank WDI 2025; IMF World Economic Outlook Oct 2025; 5-year averages 2020–2025</li><li>Healthcare: WHO World Health Statistics 2024; Numbeo Healthcare Index 2025; OECD Health at a Glance 2024. Scores calibrated for foreign postdoctoral researchers on employment contracts.</li><li>Living Wage — methodology: All figures are net monthly cost floors (what you need to spend, not earn) — enabling direct, fair comparison with postdoc net salary. US cities: MIT Living Wage Calculator 2025 “Typical Expenses” table (1 adult, 0 children) ÷ 12, which explicitly excludes income taxes. MIT’s headline living wage (gross) was NOT used. All other cities: (rent + food + transport + utilities) × 1.30 buffer, rounded to nearest $50, covering internet, clothing, personal care and misc. Japan +$250 for mandatory NHI + pension self-payment by JSPS fellows. Turkey: national benchmark retained. All USD at Feb 2026 rates.</li><li>Industry postdoc: Glassdoor, <a href="http://levels.fyi/">Levels.fyi</a>, LinkedIn salary data; representative pharma/biotech hubs 2025–2026. USD values use Feb 2026 exchange rates.</li></ul><p>All figures are estimates for comparative purposes. Verify current rates with funding bodies, institutions, and national tax authorities before making career decisions.</p><p>Hope you find it useful.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f9f06bd37835" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Kedisiz Kitaplar]]></title>
            <link>https://fatmab-dincaslan.medium.com/kedisiz-kitaplar-14c35978a517?source=rss-d035227fb6ad------2</link>
            <guid isPermaLink="false">https://medium.com/p/14c35978a517</guid>
            <dc:creator><![CDATA[Ortaya Karışık (Fatma Betul Dincaslan)]]></dc:creator>
            <pubDate>Mon, 13 Jul 2026 00:23:07 GMT</pubDate>
            <atom:updated>2026-07-13T00:25:23.216Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>Zannediyorum ki benim kadar her telden okumayı seven az. Seçici olmaya çalıştım. Umarım siz de faydalanırsınız.</blockquote><p>Son ~10 yılda (kimisini yeniden) okuduğum ve bakış açımı zenginleştirdiğini düşündüğüm bazı eserleri sıraladım.</p><ul><li>- İrlanda Defteri | Meltem Gürle <br>- Imagining the Middle East | Thierry Hentsch<br>- Isonomia and the Origins of Philosophy | Kojin Karatani<br>- Pedagogy of Freedom | Paulo Freire<br>- The Happiness Trap | Russ Harris <br>- Escape From Freedom/The Anatomy of Human Destructiveness/To Have or To Be | Erich Fromm<br>- An Enquiry Concerning Human Understanding | David Hume<br>- Strangers Gods and Monsters | Richard Kearney<br>- Man’s Wordly Goods | Leo Huberman<br>- Breaking Through | Katalin Kariko<br>- On Liberty | John Stuart Mill<br>- The Social Contract | Jean-Jacques Rousseau<br>- Nudge | Richard H. Thaler<br>- The Other Side of the Couch | Gary Small<br>- Siddartha | Hermann Hesse<br>- Milli Mücadele Tarihi | Halil İnalcık<br>- Varolmak Gelişmek Uzlaşmak | Üstün Dökmen<br>- Leo Africanus | Amin Maalouf<br>- Dokuz Yüz Katlı İnsan | Mustafa Merter<br>- Yaşar Ne Yaşar Ne Yaşamaz | Aziz Nesin<br>- Bir Türkiye Hayali/Bakışınızı Değiştirecek 10 Deney | Selçuk Şirin<br>- Always Coming Home | Ursula K. Le Guin<br>- Blindness/Seeing/The Cave | Jose Saramago<br>- İbrahim Efendi Konağı | Samiha Ayverdi<br>- Beş Şehir | Ahmet Hamdi Tanpınar<br>- The Symposium | Plato<br>- İnsan Olmak/Hayat/Psikanaliz ve Sonrası | Engin Geçtan<br>- Zoraki Diplomat | Yakup Kadri Karaosmanoğlu<br>- A Confession | Leo Tolstoy</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=14c35978a517" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[mRNA2025Berlin: From a Perspective]]></title>
            <link>https://fatmab-dincaslan.medium.com/mrna2025berlin-from-a-perspective-261706cbe624?source=rss-d035227fb6ad------2</link>
            <guid isPermaLink="false">https://medium.com/p/261706cbe624</guid>
            <dc:creator><![CDATA[Ortaya Karışık (Fatma Betul Dincaslan)]]></dc:creator>
            <pubDate>Fri, 26 Dec 2025 02:06:42 GMT</pubDate>
            <atom:updated>2025-12-26T02:06:42.035Z</atom:updated>
            <content:encoded><![CDATA[<h4><em>From RNA therapeutics and vaccine development to nanoparticle formulation and manufacturing</em></h4><p>mRNA Health Conference has a suprisingly long history. The first ever mRNA health conference (yes, before RNA got so popular!) was held in Tubingen, Germany in 2013. You can find the details and the speakers here (<a href="https://mrna-conference.com/conference-2013/#:~:text=The%201st%20International%20mRNA%20Health%20Conference%2C%20held,for%20the%20first%20time%20more%20than%20150">https://mrna-conference.com/conference-2013/#:~:text=The%201st%20International%20mRNA%20Health%20Conference%2C%20held,for%20the%20first%20time%20more%20than%20150</a>).</p><p>I participated in the 13th mRNA Health Conference (my first time) with my fantastic teammates (here is the link for the details: <a href="https://www.linkedin.com/posts/fbdincaslan_home-13th-international-mrna-health-conference-activity-7368345106493399041-oww7?utm_source=share&amp;utm_medium=member_desktop&amp;rcm=ACoAAByke68BYBET-mNwVLdKuEuEBB32RjzAYOE">https://www.linkedin.com/posts/fbdincaslan_home-13th-international-mrna-health-conference-activity-7368345106493399041-oww7?utm_source=share&amp;utm_medium=member_desktop&amp;rcm=ACoAAByke68BYBET-mNwVLdKuEuEBB32RjzAYOE</a>).</p><p>Here is the link for the conference webpage for more details: <a href="https://mrna-conference.com/">https://mrna-conference.com</a>.</p><p>The 3 days intensive conference was actually started a night earlier with a “Welcome Networking Event” at the venue.</p><p>The first day of the conference was kicked off with a keynote speech and various sessions covering therapeutics and vaccine development. We had an opportunity to learn about “Sleeping Beauty Gene Tranfer” systems, Lipid Nanoparticle (LNP) encapsulated mRNA therapeutics for asthma, therepeutics for rare metabolic diseases, targeted therapies with ionazible lipids, mRNA vs. protein vaccines, phylogenetic grouping of antigens, strategies against Clostridium Difficile (C. diff) infections, mRNA vaccines against urinary tract infections (UTI), LNPs as potential adjuvants. Furthermore, Olujimi Olaghere’s personal journey as a genetically modified patient (CRISPR-based sickle cell therapy, CTX001) was inspiring. He mentioned about making these therapies affordable for everyone. There were also round table discussions with invited speakers from US, Europe and Australia, flask talks from different companies and poster presentations of selected abstracts (odd numbered, I also presented my poster titled as “High Througput Screening for circRNA Therapeutics”). The final part of the first day was the “Networking Dinner”.</p><p>The second day began with delivery systems focused talks, and continued with RNA and vaccines. I listened to the talks covering these topics: chemical evolution of nanocarriers for RNA delivery and genome editing, novel ionizable lipids, next generation LNPs, investigating mRNA cellular stability and decay mechanisms (with TAIL-seq), self amplifying RNA (saRNA) regulation by innate immune system, snoRNAs regulating protein secretion (KARR-seq and snoKARR-Seq), identifying cellular mechanisms such as adenylation dynamics to enhace mRNA vaccination, re-evaluating sequence requirement for translation initiation and consensus sequences such as Kozak, and the potential role of extended translation initiation sites (eTIS), optimization of mRNA for higher protein expression with untranslated region (UTR) screening, potential immonogeneticy and reactogenecity of LNPs and optimization strategies and pH-cleavable alternatives, AMG lipids as an alternative to epoxide ring opening chemistry for novel LNPs, characterization of each components from efficienct encapsulation of RNA to improved endosomal escape and clearance. The day was ended with a keynote speech and poster presentations (even numbered).</p><p>The third day of the conference started with manufacturing and therapeutics and was completed with the second session of the delivery systems. There were a lot talks from companies regarding their advantage for different processes such as mRNA purity, in vitro transcription (IVT) systems, and LNPs. Nevertheless, we were lucky to listen to the developers of personalised gene-editing therapy (CRISPR for CPS1 deficiency, a rare autosomal recessive disorder) that cured a new-born baby (baby KJ) in short period of time. After CureVac passed the baton Moderna, the conference was ended.</p><p>I was curious about the diversity of participants. Although I am not sure about the exact numbers (which is only available to conference organisers. And please careful affiliations do not reflect the nationalities), the networking page of the conference platform gives an idea. Europe (EU) participants were affiliated with companies or institutions overwhelmingly from Germany (=&gt;50% among EU, and =&gt;25% in total) and the conference was, not suprisingly, dominated by EU (including UK) and North America (USA and Canada) (~80% in total). Yet, there were a broad interest from East Asia (China, Japan, South Korea, making &gt;10%). Although, there were a few representatives from other regions such as Latin America, South Africa and Australia, I think it might increase in the next few years. Aside from visa related hesitations to attend the conference, this might be indicating the interest of regions taking part in current trends in biotech research and innovation as well.</p><p>I learnt a lot during the conference. Moreover, I had a chance to meet one of the brilliant minds of our century: Katalin Kariko, and Biontech representatives. Pro tip: Make sure you read Dr. Kariko’s book before introducing yourself!</p><p>The organisers (and sponsors) provided snacks, drinks, and a platform to exchange our knowledge about recent advacements in RNA therepeutics, delivery systems, vaccine development and manufacturing (All joking aside, we were suprized and delighted with the diversity of vegan and vegetarian options).</p><p>Thanks a lot for this wonderful event and the opportunity to visit one of the most vibrant cities in Germany!</p><p>Hope to see more of life-saving and affordable therapies for everyone and increasing diversity of participants all over the world.</p><p>#mRNA #therapeutics #Berlin2025 #mRNAHealthConference #nanoparticle #deliverysystems #RNAtherapeutics</p><p>If you are a PhD student or post-doc with accepted poster, the conference is free (you do not need to pay for the registration fee)!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=261706cbe624" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Academic Funding and Fellowship Opportunities]]></title>
            <link>https://fatmab-dincaslan.medium.com/funding-and-fellowship-opportunities-in-ireland-513ba0b30488?source=rss-d035227fb6ad------2</link>
            <guid isPermaLink="false">https://medium.com/p/513ba0b30488</guid>
            <dc:creator><![CDATA[Ortaya Karışık (Fatma Betul Dincaslan)]]></dc:creator>
            <pubDate>Wed, 24 Dec 2025 21:41:54 GMT</pubDate>
            <atom:updated>2025-12-24T22:00:43.445Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>I wrote it <a href="https://www.linkedin.com/pulse/funding-fellowship-opportunities-ireland-dincaslan-phd--h9ycc">here</a>. The brilliant team in UCC shares a weekly newsletter of research funding opportunities in Ireland. As a celebration of my first year of post-doc in Ireland, I decided to bring all the relevant links for international (including non-EU) researchers [also some for my future self].</blockquote><p>[Although some of these provide openings for a predoc/PhD as well, the intended audience is PhD holders: Post-Doc, ECR.]</p><p>If you are a life scientist looking for opportunities after a PhD and are unsure where/how to start in Ireland, we’re here for you!</p><p>Please keep in mind that most of the EU postdoctoral fellowships require mobility. This means you cannot apply for them after a while from the same lab/country as you need to move to! Unfortunately, some of them have very strict time requirements, such as within 3 years of a PhD degree, to be eligible. It would be better to apply them at the beginning of your job.</p><h3>Postdoctoral Fellowships</h3><ul><li>Government of Ireland [Postdoctoral]/Research Ireland Fellowship Programme, <a href="https://www.researchireland.ie/funding/government-ireland-postdoctoral/">https://www.researchireland.ie/funding/government-ireland-postdoctoral/</a></li><li>EMBO [Postdoctoral] Fellowship, <a href="https://www.embo.org/funding/fellowships-grants-and-career-support/postdoctoral-fellowships/">https://www.embo.org/funding/fellowships-grants-and-career-support/postdoctoral-fellowships/</a></li><li>MSCA Fellowships [Postdoctoral/ECR], <a href="https://marie-sklodowska-curie-actions.ec.europa.eu/calls/msca-postdoctoral-fellowships-2025">https://marie-sklodowska-curie-actions.ec.europa.eu/calls/msca-postdoctoral-fellowships-2025</a> and <a href="https://marie-sklodowska-curie-actions.ec.europa.eu/calls/msca-staff-exchanges-2025">https://marie-sklodowska-curie-actions.ec.europa.eu/calls/msca-staff-exchanges-2025</a></li><li>Research Ireland Enterprise [PostDoc] Fellowship, <a href="https://www.researchireland.ie/funding/enterprise-fellowships/">https://www.researchireland.ie/funding/enterprise-fellowships/</a></li><li>L’Oreal for Women in Science [non-profit/private, PostDoctoral/ECR], <a href="https://www.forwomeninscience.com/challenge/show/142">https://www.forwomeninscience.com/challenge/show/142</a></li></ul><h3>Early Career Researcher [ECR] Grants/Funding for Research</h3><ul><li>Royal Society-Research Ireland Fellowship [Transition from postdoc to group leader], <a href="https://www.researchireland.ie/funding/royal-society-university/">https://www.researchireland.ie/funding/royal-society-university/</a> and <a href="https://royalsociety.org/grants/university-research/">https://royalsociety.org/grants/university-research/</a></li><li>Welcome Trust Eary Career Awards [Transition from postdoc to group leader], <a href="https://wellcome.org/research-funding/schemes/wellcome-early-career-awards">https://wellcome.org/research-funding/schemes/wellcome-early-career-awards</a></li><li>ERC Starting Grant, <a href="https://erc.europa.eu/apply-grant/starting-grant">https://erc.europa.eu/apply-grant/starting-grant</a></li><li>HRB Grants and joint calls, <a href="https://www.hrb.ie/funding-category/research-funding/funding-opportunities/grant-schemes/">https://www.hrb.ie/funding-category/research-funding/funding-opportunities/grant-schemes/</a> (e.g., <a href="https://www.hrb.ie/funding-scheme/investigator-led-projects-ilp-2026/">https://www.hrb.ie/funding-scheme/investigator-led-projects-ilp-2026/</a>)</li><li>Global Talent Ireland [ECR/Senior], <a href="https://www.researchireland.ie/funding/global-talent-ireland/">https://www.researchireland.ie/funding/global-talent-ireland/</a></li><li>EMBO Young Investigator Programme, <a href="https://www.embo.org/funding/fellowships-grants-and-career-support/young-investigator-programme/application/">https://www.embo.org/funding/fellowships-grants-and-career-support/young-investigator-programme/application/</a></li><li>Lister Institute Prize [non-profit/private, Transition from postdoc to group leader], <a href="https://lister-institute.org.uk/">https://lister-institute.org.uk/</a></li></ul><h3>Exchange Programs for short duration</h3><ul><li>EMBO Scientific Exchange Grant, <a href="https://www.embo.org/funding/fellowships-grants-and-career-support/scientific-exchange-grants/">https://www.embo.org/funding/fellowships-grants-and-career-support/scientific-exchange-grants/</a></li><li>MSCA Staff Exchange Programme, <a href="https://marie-sklodowska-curie-actions.ec.europa.eu/calls/msca-staff-exchanges-2025">https://marie-sklodowska-curie-actions.ec.europa.eu/calls/msca-staff-exchanges-2025</a></li></ul><h3>More of these for everyone</h3><ul><li>Research Ireland programs, <a href="https://www.researchireland.ie/funding/">https://www.researchireland.ie/funding/</a></li><li>European Molecular Biology Organization (EMBO), <a href="https://www.embo.org/funding/fellowships-grants-and-career-support/">https://www.embo.org/funding/fellowships-grants-and-career-support/</a></li><li>Welcome Trust, <a href="https://wellcome.org/research-funding/schemes">https://wellcome.org/research-funding/schemes</a></li><li>European Research Council (ERC), <a href="https://erc.europa.eu/apply-grant">https://erc.europa.eu/apply-grant</a></li><li>European Innovation Council (EIC), <a href="https://eic.ec.europa.eu/eic-funding-opportunities_en">https://eic.ec.europa.eu/eic-funding-opportunities_en</a></li><li>Marie Sklodowska Curie Actions (MSCA), <a href="https://marie-sklodowska-curie-actions.ec.europa.eu/actions">https://marie-sklodowska-curie-actions.ec.europa.eu/actions</a></li><li>Health Research Board (HRB) and joint calls, <a href="https://www.hrb.ie/funding-category/research-funding/funding-opportunities/grant-schemes/">https://www.hrb.ie/funding-category/research-funding/funding-opportunities/grant-schemes/</a></li><li>EMBL ARISE Fellowship, <a href="https://www.embl.org/about/info/postdoctoral-programme/fellowship-programmes/">https://www.embl.org/about/info/postdoctoral-programme/fellowship-programmes/</a> [a partner from an EMBL Member State]</li><li>Horizon Europe 2025 details in general, <a href="https://research-and-innovation.ec.europa.eu/funding/funding-opportunities/funding-programmes-and-open-calls/horizon-europe/horizon-europe-work-programmes_en">https://research-and-innovation.ec.europa.eu/funding/funding-opportunities/funding-programmes-and-open-calls/horizon-europe/horizon-europe-work-programmes_en</a></li></ul><p>And many more… Some examples:</p><ul><li>Science communication, <a href="https://www.researchireland.ie/funding/science-week-call-2025/">https://www.researchireland.ie/funding/science-week-call-2025/</a></li><li>Charitable funding opportunities such as for T1D, <a href="https://www.breakthrought1d.org/explore-research/funding-opportunities/">https://www.breakthrought1d.org/explore-research/funding-opportunities/</a></li><li>Participating Lindau Nobel meetings, <a href="https://www.lindau-nobel.org/young-scientists/">https://www.lindau-nobel.org/young-scientists/</a></li><li>Conference organization support, <a href="https://www.nui.ie/awards/PostDoctoralFellowships/#Early_Career_Academics">https://www.nui.ie/awards/PostDoctoralFellowships/#Early_Career_Academics</a></li></ul><p>If you are interested in research positions in the EU in general, I highly recommend checking out the EURAXESS page as well: <a href="https://euraxess.ec.europa.eu/jobs/search">https://euraxess.ec.europa.eu/jobs/search</a>.</p><p>If you decide to continue your fantastic research in Ireland, it is better not to miss the EU joint calls.</p><p>And please keep in mind that most of these calls still prioritize EU applicants over non-EU applicants.</p><p>All the best, pal!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=513ba0b30488" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Scribbles from Essentials of AI]]></title>
            <link>https://fatmab-dincaslan.medium.com/scribbles-from-essential-of-ai-2726ff3a378d?source=rss-d035227fb6ad------2</link>
            <guid isPermaLink="false">https://medium.com/p/2726ff3a378d</guid>
            <dc:creator><![CDATA[Ortaya Karışık (Fatma Betul Dincaslan)]]></dc:creator>
            <pubDate>Wed, 24 Dec 2025 21:36:19 GMT</pubDate>
            <atom:updated>2025-12-24T22:08:22.590Z</atom:updated>
            <content:encoded><![CDATA[<blockquote><em>I wrote it </em><a href="https://www.linkedin.com/pulse/scribbles-from-essential-ai-fatma-betul-dincaslan-phd-hxrhc"><em>here</em></a><em>. UCC launched an artificial intelligence (AI) course “for everyone” this year: “</em><strong><em>UW0013: Essentials of AI</em></strong><em>”. The course was given by Dr. Barry O’Sullivan. </em>I completed several Machine Learning (ML) courses before. I also kept reading about AI; however, I would prefer a structured ecosystem to learn the regulations and updates from a European perspective on AI. This course was an excellent opportunity, so I signed up.</blockquote><p>We had 9 evening classes. You could participate in person or online. I usually joined while running late evening experiments, basic analysis, after doing sports, or cooking a meal. Not surprisingly, I need to pay more attention before the quiz and take some notes to grasp the concepts.</p><h4><em>Please remember that AI is not a human-like-machine that thinks; it is rather a set of algorithms/tools that developers and managers decide how to train, apply, and proceed. The bias or benefits all come from the intention of the people who develop, deploy, or use.</em></h4><p>Here is some useful information for you as well.</p><h3>1) An Overview of AI</h3><p>In case you have not discovered yet, Dr. O’Sullivan shared a free website to learn the <em>basics of ML and AI:</em> <a href="https://course.elementsofai.com/en-ie/">https://course.elementsofai.com/en-ie/</a>.</p><p>He introduced the fundamental concepts such as what AI is, Natural Language Processing (NLP), and ChatGPT are, the energy consumption of these tools, traditional programming vs Machine Learning (ML) vs Deep Learning, some elementary concepts like “<em>bag of words (BoWs)</em>” representation, the implementation of AI in Sustainable Development Plans, and <strong>European Union AI Act </strong>(<a href="https://artificialintelligenceact.eu/">https://artificialintelligenceact.eu</a>), how <em>Generative AI </em>(GenAI) systems as plagiarism by design, hallucinations, ups, hypes and downs of AI, military use and ethical issues, intellectual property and confidentiality, AI and education.</p><h3>2) Some Methods and Topics in AI</h3><p>The lecture started with the branches of AI, from ML, NLP, and planning to vision, speech, and robotics. We learnt that the EU AI Act covers “<strong>AI system</strong>” by definition, rather than some GenAI-focused material or algorithm-focused definitions.</p><p>The reason was simply that not all AI is only about data or GenAI, but also about logics, modeling, and the automation of reasoning. That brought us to “Constraint Satisfaction and Constraint Programming.” He also explained the “Recommender Systems” used in Netflix-like platforms.</p><p>We discussed ML in three main topics: “<strong>Classification</strong>”, such as OneR, holdout estimations, or leave-one-out validations, and k-fold cross-validations, to evaluate the AI systems and minimize the mistakes. “<strong>Clustering</strong>” such as Naïve Bayes classifier thinking frequencies as possibilities, or Decision Trees, contingency tables, and biases in ML applications, GenAI systems, and “<strong>Regression</strong>”.</p><p>As we are not living in an equal world, the bias can be minimized; however, complete elimination is not possible.</p><p>Here are some interesting resources shared for the lecture:</p><ul><li>Addressing sources of bias in culture, including technology <a href="https://www.science.org/doi/10.1126/science.aal4230">https://www.science.org/doi/10.1126/science.aal4230</a></li><li>Understanding the Cambridge Analytica scandal, <a href="https://www.amazon.co.uk/Outnumbered-Facebook-Filter-bubbles-Algorithms-Control/dp/147294741X">https://www.amazon.co.uk/Outnumbered-Facebook-Filter-bubbles-Algorithms-Control/dp/147294741X</a></li></ul><h3>3) The Ethics of AI</h3><p>We began with the trolley dilemma for machines and the hurdles of international perception of ethics.</p><p>It turns out that there are 7 key requirements for <strong>Trustworthy AI</strong> to achieve ethical standards across Europe (please feel free to check it out, <a href="https://digital-strategy.ec.europa.eu/en/library/ethics-guidelines-trustworthy-ai">https://digital-strategy.ec.europa.eu/en/library/ethics-guidelines-trustworthy-ai</a>).</p><p>We learnt that there are four main risks associated with AI systems in the EU used in regulatory schemes and governance across the EU (<a href="https://www.euaiact.com/key-issue/3#:~:text=The%20EU%20AI%20Act%20introduces,health%2C%20safety%20and%20fundamental%20rights">https://www.euaiact.com/key-issue/3#:~:text=The%20EU%20AI%20Act%20introduces,health%2C%20safety%20and%20fundamental%20rights</a>.):</p><p>I) <strong>Unacceptable risk </strong>that prohibits such activities (e.g., social scoring, scrapping internet or CCTV for images for databases),</p><p>II) <strong>Higher risks</strong> with stringent obligations (e.g., CV-sorting software for recruitment, critical infrastructure),</p><p>III) <strong>Limited risks </strong>with transparency requirements (e.g., impersonation bots),</p><p>IV) <strong>Minimal risk </strong>without restriction but code of conducts to be encouraged.</p><p>The course covered the military aspects of AI applications. It seemed some of the weapons were actually prohibited (not sure if it applies to everyone).</p><p>The act about the risks provides four main dimensions of AI usage: I) Design and Development, II) Deployment and Use with Responsibility, III) Post-Use assessment and Accountability, IV) Sharing, Trust building, and Cooperation with Transparency.</p><p>Here are some interesting resources shared for the lecture:</p><ul><li>The Moral Machine <a href="https://www.nature.com/articles/s41586-018-0637-6">https://www.nature.com/articles/s41586-018-0637-6</a></li><li>OECD Principles of AI <a href="https://oecd.ai/en/ai-principles">https://oecd.ai/en/ai-principles</a></li><li>Code of Conduct for Humanitarian Use of AI in military settings <a href="https://hdcentre.org/news/release-of-draft-code-of-conduct-on-ai-enabled-military-systems/">https://hdcentre.org/news/release-of-draft-code-of-conduct-on-ai-enabled-military-systems/</a></li></ul><h3>4) AI, Skills, and Future of Work</h3><p>This section is for those out there who are scared that they will lose their job due to AI. Fret not!</p><p>The bad news is that, unfortunately, some jobs may be lost. On the other hand, the good news is that there will be more jobs to be created. The real challenge will be the fair sharing of the growth among everyone involved in the process, as always happens in new economic shifts. Therefore, we discussed the future of jobs in the age of AI, the real adoption rate of AI in the sectors, the skillsets for better preparedness, and the social contracts to provide humane standards for everyone.</p><p>We love exaggerating the technology and development all the time. However, we haven’t seen the flying cars yet. It seems the adoption of AI in the sectors was not as fast as people predicted or advertised. So, Barry reminded us of <em>(Roy) Amara’s Law (1978)</em> about it: “<strong>We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.</strong>”. The example about social media and influencers as a job was interesting and captivating.</p><p>As <em>(Arthur) Okun’s Law (1962) </em>suggests, <strong>growth happens with employment</strong>. If you are a lifelong learner and analytical thinker, stay agile and resilient, not afraid of gaining new skills in AI and technology, then you are future-ready.</p><p>Here are some interesting resources shared for the lecture:</p><ul><li>The Future of Jobs, WEF <a href="https://www.weforum.org/publications/the-future-of-jobs-report-2025/digest/">https://www.weforum.org/publications/the-future-of-jobs-report-2025/digest/</a></li><li>AI-ready future skills <a href="https://enterprise.gov.ie/en/publications/ai-skills.html">https://enterprise.gov.ie/en/publications/ai-skills.html</a></li><li>Jobs lost, jobs gained <a href="https://www.mckinsey.com/featured-insights/future-of-work/jobs-lost-jobs-gained-what-the-future-of-work-will-mean-for-jobs-skills-and-wages#/">https://www.mckinsey.com/featured-insights/future-of-work/jobs-lost-jobs-gained-what-the-future-of-work-will-mean-for-jobs-skills-and-wages#/</a></li><li>Universal Basic Income as a New Social Contract in the age of AI <a href="https://blogs.lse.ac.uk/businessreview/2025/04/29/universal-basic-income-as-a-new-social-contract-for-the-age-of-ai-1/">https://blogs.lse.ac.uk/businessreview/2025/04/29/universal-basic-income-as-a-new-social-contract-for-the-age-of-ai-1/</a></li><li>AI adoption in the global workforce <a href="https://www.goldmansachs.com/insights/articles/how-will-ai-affect-the-global-workforce">https://www.goldmansachs.com/insights/articles/how-will-ai-affect-the-global-workforce</a></li><li>AI Job Risk Calculator, <a href="https://aijobrisk.in/">https://aijobrisk.in/</a> and <a href="https://lis2.epfl.ch/resiliencetorobots/#/">https://lis2.epfl.ch/resiliencetorobots/#/</a></li><li>Find A Career based on Interest and Skills <a href="https://www.onetonline.org/">https://www.onetonline.org/</a></li></ul><h3>5) AI and Education</h3><p>AI is a handy tool in/for education and personalized learning. We discussed this in two main aspects: <strong>AI for education </strong>and <strong>AI in education</strong>.</p><p>He introduced us to the areas in which AI can be implemented in the educational frameworks, from adaptive learning to intelligent assessments.</p><p>Moreover, the experts shared guidelines about the ethical use of AI in educational settings in UCC and Ireland.</p><p>An <em>AI Literacy</em> course is important in this regard for the responsible use of AI. For example, you are required to acknowledge the use of AI in UCC with transparent sharing of the specific prompts.</p><p>Here are some interesting resources shared for the lecture:</p><ul><li>AI on learning (potential risks) <a href="https://www.newyorker.com/culture/infinite-scroll/ai-is-homogenizing-our-thoughts">https://www.newyorker.com/culture/infinite-scroll/ai-is-homogenizing-our-thoughts</a> and <a href="https://www.brainonllm.com/">https://www.brainonllm.com/</a></li><li>AI in education <a href="https://www.sciencedirect.com/science/article/pii/S0957417424010339">https://www.sciencedirect.com/science/article/pii/S0957417424010339</a></li><li>Framework for ethical integration of AI (no AI to full GenAI tasks <a href="https://open-publishing.org/journals/index.php/jutlp/article/view/810/769">https://open-publishing.org/journals/index.php/jutlp/article/view/810/769</a></li><li>UCC Toolkit for Ethical Use of GenAI in teaching and learning <a href="https://www.ucc.ie/en/ethical-use-of-generative-ai-toolkit/">https://www.ucc.ie/en/ethical-use-of-generative-ai-toolkit/</a>, and <a href="https://www.ucc.ie/en/ethical-use-of-generative-ai-toolkit/academic-integrity/">https://www.ucc.ie/en/ethical-use-of-generative-ai-toolkit/academic-integrity/</a></li><li>Pros and Cons of AI in Education <a href="https://pce.sandiego.edu/ai-impact-education/">https://pce.sandiego.edu/ai-impact-education/</a></li><li>Higher Education Authority Guidelines of AI in Ireland <a href="https://hea.ie/2025/12/22/hea-publishes-national-policy-framework-on-generative-ai-in-teaching-and-learning/">https://hea.ie/2025/12/22/hea-publishes-national-policy-framework-on-generative-ai-in-teaching-and-learning/</a></li></ul><h3>6) AI, Media and Democracy</h3><p>It is already a challenging task to distinguish AI-written text from well-written human texts. However, the audio and deepfake are even more challenging aspects of AI. The Perplexity Fake Podcast is a good example given during the class.</p><p>There was an interesting application of AI in journalism, which is predicting the trending news using Text Classification and “<em>Topic Modeling</em>” out of BoWs. Storyful was an interesting example applying it provide actionable digital media intelligence.</p><p>However, LLMs in a personalized setting are considered a double-edged sword. It raises ethical concerns regarding microtargeting, such as marketing and manipulating users, as happened in some of the elections before. Another issue was Digital Authoritarianism, with decreased freedom on the internet.</p><p>It turns out that GenAI has great potential to become a disinformation amplifier. Unfortunately, the vulnerable group affected by all this nonsense is mostly women or minorities. All these brought us the importance of regulation and governance, international standards, coordinated response, education, and training regarding AI across the globe.</p><p>Here are some interesting resources shared for the lecture:</p><ul><li>Lies spread faster than truth online <a href="https://www.science.org/doi/10.1126/science.aap9559">https://www.science.org/doi/10.1126/science.aap9559</a></li><li>Storyful <a href="https://storyful.com/">https://storyful.com/</a></li><li>Information Ecosystem and Troubled Democracy <a href="https://observatory.informationdemocracy.org/wp-content/uploads/2025/06/rapport_forum_information_democracy_2025-1.pdf">https://observatory.informationdemocracy.org/wp-content/uploads/2025/06/rapport_forum_information_democracy_2025-1.pdf</a></li><li>Fake Content Generation is Immense <a href="https://journal.everypixel.com/ai-image-statistics">https://journal.everypixel.com/ai-image-statistics</a></li></ul><h3>7) AI and Sustainability</h3><p>Surprisingly or not surprisingly, AI has already been implemented in the 17 SDGs. Towards a similar aim, researchers around the world have been applying AI, such as a combination of satellite imaging and deep learning, as low-cost and globally available tools to predict and prevent poverty.</p><p>Another application discussed throughout the lecture was energy cost minimization of mining using constraint programming. The Google-AI Green Light project was also a good application of AI, reducing traffic emissions to build resilient and sustainable cities. The application of terrestrial ecosystems to manage forests and reduce biodiversity loss was a terrific idea.</p><p>We also discussed the hidden costs of AI from natural resources (e.g., carbon footprints) and privacy to freedom and equity.</p><p>Here are some interesting resources shared for the lecture:</p><ul><li>Energy consumption and sustainability issues <a href="https://epoch.ai/gradient-updates/how-much-energy-does-chatgpt-use#:~:text=We%20find%20that%20typical%20ChatGPT,less%20than%20the%20older%20estimateand">https://epoch.ai/gradient-updates/how-much-energy-does-chatgpt-use#:~:text=We%20find%20that%20typical%20ChatGPT,less%20than%20the%20older%20estimateand</a> <a href="https://medium.com/illuminations-mirror/is-chatgpt-the-next-energy-guzzler-f99454e53c62">https://medium.com/illuminations-mirror/is-chatgpt-the-next-energy-guzzler-f99454e53c62</a></li><li>United Nations (UN) 17 Sustainable Development Goals (SDGs) <a href="https://sdgs.un.org/goals">https://sdgs.un.org/goals</a></li><li>Emissions based on infrastructure and power consumption <a href="https://codecarbon.io/">https://codecarbon.io/</a>, ML Emission calculator <a href="https://mlco2.github.io/impact/#compute">https://mlco2.github.io/impact/#compute</a> , AI carbon footprint calculator <a href="https://www.deloitte.com/uk/en/services/consulting/content/ai-carbon-footprint-calculator.html">https://www.deloitte.com/uk/en/services/consulting/content/ai-carbon-footprint-calculator.html</a></li></ul><h3>8) AI and Health and Medicine</h3><blockquote>If we exclude the possible privacy concerns, I think one of the best uses of AI is in health and medicine. It shows a great promise for the tool.</blockquote><p>Dr. Helmut Simonis shared examples of AI applications in health domains. These were Nurse Rostering, UCC Dental School Timetabling, Outpatient Wait-list management, and Integrated Hospital Management. He demonstrated the fundamentals of modeling, optimization, decision making, and constraint programming in real-life scenarios to solve combinatorial problems.</p><p>Dr. Andrea Visentin showed case studies of AI on medical application including Raman spectroscopy-based detection, surgical guide systems, lactate measurement, and even speech therapy (e.g., EaseTalk). He also emphasized the <strong>data quality</strong> and <strong>explainability</strong> for clinical trust.</p><p>Here are some interesting resources shared for the lecture:</p><ul><li>Principles for artificial intelligence (AI) and its application in healthcare in UK <a href="https://www.bma.org.uk/advice-and-support/nhs-delivery-and-workforce/technology/principles-for-artificial-intelligence-ai-and-its-application-in-healthcare">https://www.bma.org.uk/advice-and-support/nhs-delivery-and-workforce/technology/principles-for-artificial-intelligence-ai-and-its-application-in-healthcare</a></li><li>The Future of AI-enabled Health <a href="https://reports.weforum.org/docs/WEF_The_Future_of_AI_Enabled_Health_2025.pdf">https://reports.weforum.org/docs/WEF_The_Future_of_AI_Enabled_Health_2025.pdf</a></li></ul><h3>9) AI and Gender</h3><p>Gender issues in AI seem ubiquitous. As stated by UNESCO, AI has a huge power to shape the perception of many people, so it might amplify the inequalities around the world as well.</p><p>It was addressed why women have less representation despite having more qualifications in certain sectors and fields. Interestingly, even the LinkedIn algorithms might have been updated to implement such biases recently to silence certain types of voices.</p><p>To tackle AI and gender issues, the perspectives of <strong>Feminist AI </strong>and <strong>data feminism</strong> were discussed. These views suggest understanding how the world operates, challenging hierarchies to promote EDI while considering the contexts carefully and making your labor visible and algorithms inclusive.</p><p>On the other hand, he mentioned that <em>data ecofeminism</em> approaches the issue from a socio-ecological view. Basically, it covers the climate emergency with power structures, and prioritizes frugal AI computing.</p><p>Here are some interesting resources shared for the lecture:</p><ul><li>Where are the women <a href="https://www.turing.ac.uk/sites/default/files/2021-03/where-are-the-women_public-policy_full-report.pdf">https://www.turing.ac.uk/sites/default/files/2021-03/where-are-the-women_public-policy_full-report.pdf</a></li><li>Voices of Her (Article) <a href="https://arxiv.org/abs/2305.14597">https://arxiv.org/abs/2305.14597</a></li><li>BroBoost on LinkedIn <a href="https://www.theguardian.com/technology/2025/nov/22/bro-boost-women-find-linkedin-traffic-drives-if-they-pretend-to-be-men">https://www.theguardian.com/technology/2025/nov/22/bro-boost-women-find-linkedin-traffic-drives-if-they-pretend-to-be-men</a></li><li>Invisible Women (Book) <a href="https://www.amazon.co.uk/Invisible-Women-Exposing-World-Designed/dp/1784741728">https://www.amazon.co.uk/Invisible-Women-Exposing-World-Designed/dp/1784741728</a></li><li>Data Feminism (Book) <a href="https://data-feminism.mitpress.mit.edu/">https://data-feminism.mitpress.mit.edu/</a></li></ul><p>I think it was an informative and well-designed course to learn about the fundamental aspects of AI, its applications, ongoing issues, and regulations. I am grateful to UCC for providing such opportunities for us free of charge.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2726ff3a378d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[FoG2025London: From a Perspective]]></title>
            <link>https://fatmab-dincaslan.medium.com/fog2025london-from-a-perspective-1e18f7f8bd26?source=rss-d035227fb6ad------2</link>
            <guid isPermaLink="false">https://medium.com/p/1e18f7f8bd26</guid>
            <category><![CDATA[big-data]]></category>
            <category><![CDATA[conference]]></category>
            <category><![CDATA[fog2025london]]></category>
            <category><![CDATA[biodata]]></category>
            <category><![CDATA[festivalofgenomics]]></category>
            <dc:creator><![CDATA[Ortaya Karışık (Fatma Betul Dincaslan)]]></dc:creator>
            <pubDate>Thu, 30 Jan 2025 23:36:46 GMT</pubDate>
            <atom:updated>2025-12-24T22:22:47.142Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>I will be sharing my notes and summaries from the talks I attended at the conference (before they start to be embedded somewhere less accessible in my memory about the event). Shared <a href="https://www.linkedin.com/pulse/fog2025london-from-perspective-fatma-betul-dincaslan-phd--z5zjc/?trackingId=vswFxftIT66pkjMUwMZ8jA%3D%3D">here</a> first.</blockquote><h3>What Was About It (#FoG2025London)</h3><p>It covers many aspects of drug discovery/development, genome-editing, rare diseases, single-cell, microbes, sequencing, and transcript/epigen/gen/prote/metabol/metagen/pharmacogen/spatial/multi-omics generating/storing/managing/using/distributing the big data. You can find more details on the event webpage (<a href="https://festivalofgenomics.com/london/en/page/2025-homepage">https://festivalofgenomics.com/london/en/page/2025-homepage</a>).</p><p>As you might guess from the topic range and industrial partnerships, the talks are usually prepared for a broader audience and are less detailed (if you know ISCB, HIBIT-like computational biology-focused conferences, it is less intense in terms of the computational nitty gritty side).</p><p>Moreover, there are a lot of companies sponsoring the event and making it free for academic researchers. Therefore, you must also be prepared for product application-oriented talks (which might be useful if you are looking for a certain type of product for your research and follow the trends with improved techniques).</p><p>Apart from this, I appreciated the clinical focus and patient-oriented talks (from patients or take-carers) that reminded (curiosity-driven) us of the most valuable aspect of biology research and encouraged us to make an invaluable difference in life.</p><p>Tip: If it is your first time attending such an event, be careful about time management. You cannot be in multiple places at the same time. Whether it is a minute decision or long-planning, take a pen or open an event planner on the phone and choose the sessions based on your interest. If only the title of the talks given, then try to google the speaker’s name and see whether the background and title matching are relevant for you. There might be a recording available, however, if you have a genuine interest and some of your questions are to be answered, give priority to that one.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*NIDOZeJA_OaxXifY" /><figcaption>ExCeL</figcaption></figure><p>I will share mostly the research/technical-oriented perspective of mine (from selected talks I attended) here.</p><h3>Day 1</h3><p><strong>Constellation: Revolutionising The Approach to Rapid Genome Sequencing in Acutely Unwell Children</strong></p><p>It covered the clinical applications of rapid-genome sequencing. Illumina representative described the PCR-free 3-step Whole Genome Sequencing (WGS) approach. For more details, you can read the product page: <a href="https://www.illumina.com/content/dam/illumina-marketing/documents/products/appnotes/illumina-dna-pcr-free-wgs-app-note-770-2020-006.pdf">https://www.illumina.com/content/dam/illumina-marketing/documents/products/appnotes/illumina-dna-pcr-free-wgs-app-note-770-2020-006.pdf</a></p><p>He also mentioned the game-changer effect of the “constellation technology” based on the recovery of ambiguous reads to improve read mapping. You can find the details here; <a href="https://www.illumina.com/science/genomics-research/articles/constellation-mapped-read-technology.html">https://www.illumina.com/science/genomics-research/articles/constellation-mapped-read-technology.html</a></p><p><strong>Putting The UK at The Forefront of Genomics and Precision Medicine</strong></p><p>It is good to know NHS Southwest Genome Medicine Service can check &gt;6000 rare conditions in a test.</p><p>Illumina mentioned how the constellation tech could be useful for various conditions here (e.g., a patient with only single parent genome availability, a genome stored &gt;10 years ago).</p><p><strong>Advancing Genomics &amp; Health Through Data</strong></p><p>Tim Hubbard gave an introduction to ELIXIR (<a href="https://elixir-europe.org/">https://elixir-europe.org</a>, e.g., database infrastructure, training programs for collaborators, establishing strategies, diverse species, and federated structure for each country for European Genome Archive-EGA to eliminate data-privacy concerns across borders).</p><p><strong>Enhancing the Nextflow Experience: Boosting Efficiency, Scalability, and Best Practices for Bioinformatics Workflows</strong></p><p>Geraldine’s talk was helpful for newcomers to pipeline development and understanding the effective role of the LLM revolution (a.k.a. some popular AI chatting tool algorithms) for faster processing. (more info: <a href="https://www.nextflow.io/">https://www.nextflow.io</a>)</p><p>She started with the challenges of good pipeline frameworks of big biodata, such as dealing with large files, the requirement of multiple coding languages, and complex algorithmic tasks. Then, she continued with the emphasis on readability, stability, portability, and reproducibility and improved these by simply using nf-core pipelines. (more info: NextFlow open-source community: <a href="https://nf-co.re/">https://nf-co.re</a>).</p><p>She mentioned the LLM-integrated AI assisting Nexflow System (NextFlow AI assistant: <a href="https://seqera.io/nextflow/">https://seqera.io/nextflow/</a>) that not only trained based on solid nf-core pipelines but also has an automated testing/debugging and correcting feature for the relevant pipeline of interest. (for those interested, you can check out the VScode extension for hands-on coding as well)</p><p><strong>Accelerating Drug Discovery Through the Power of Genomics, Machine Learning, and Medical Big Data</strong></p><p>Boehringer Ingelheim’s representative shared examples of how to effectively use machine learning (ML)/deep learning based on large-scale, multi-modal, and longitudinal genomic data for drug discovery.</p><p><strong>Identification and Validation of Non-Coding Pharmacological Target</strong></p><p>AstraZeneca’s perspective of using non-coding targets and taking advantage of them (e.g., low abundance requiring low dose treatments) was interesting. Although there were structure and expression challenges in targeting non-coding, they took advantage of interactions (such as RNA-DNA) for identification in the proximity (more info about RADICL-seq; <a href="https://www.nature.com/articles/s41467-020-14337-6">https://www.nature.com/articles/s41467-020-14337-6</a>).</p><p>NAT antisense transcripts in the context of therapeutics using assisting oligo (ASO) were discussed.</p><p><strong>Designing Efficient, Manufacturable &amp; Safe mRNA Therapeutics</strong></p><p>Tobias von der Haar discussed the different usage of mRNA formulations (e.g., antigens, antibodies, secreted proteins, or intracellular proteins) and how to design mRNA-based therapeutics for this considering yield, manufacturability, off-target effects with modifying Untranslated Region (UTR) and ORF (Open Reading Frame).</p><p><strong>Integrative Analysis of RNA-Protein Interactome Dynamics with Different Datasets Reveals Enzymatic Insights</strong></p><p>Dr. Guber mentioned two aspects of RNA-protein interaction research: RNA or protein-centric. Protein-centric approaches were summarized as RIP, CLIP, PAR-CLIP. RIP was used as an RNA-centric approach. Particularly, mRNA-binding RNA Binding Proteins (RBPs) were discussed (a.k.a. mRBPs).</p><p>He mentioned that these mRBPs are evolutionarily conserved, pointing to functional importance across species (more info: <a href="https://pubmed.ncbi.nlm.nih.gov/26595419/">https://pubmed.ncbi.nlm.nih.gov/26595419/</a>). Further investigations revealed that there is a competition between mRBPs and enzymatic activity, indicating mRBPs’ role in modulating fast response to change.</p><h3>Day 2</h3><p><strong>The Rare Disease Multiomics Programme at Genomics England</strong></p><p>I attended this talk to learn more about rare disease programs. However, I learned more than that! Although I had been using Illumina sequencing (such as Miseq) during my PhD, I realized that I was not aware of some of their products in other omics stages such as SOMAmer technology for proteomics (more info: <a href="https://www.illumina.com/products/by-type/sequencing-kits/library-prep-kits/protein-prep.html">https://www.illumina.com/products/by-type/sequencing-kits/library-prep-kits/protein-prep.html</a>).</p><p>SOMAmer is oligonucleotides folding into different shapes/secondary structures binding to proteins. It looks like they are using barcoded SOMAmers for the identification of unique proteins up to ~6k (for now) following sequencing for easy workflow. If you are looking for a Mass Spectrometry (MS) alternative for high throughput (HT) protein detection, I think it is worth checking it out.</p><p><strong>Enabling Innovation with FAIR OMICS Data Management: A Boehringer Ingelheim Perspective</strong></p><p>I think Vitaly’s talk was very well delivered. He mentioned how costly and time-consuming it is not to be “FAIR” (<a href="https://www.go-fair.org/fair-principles/">https://www.go-fair.org/fair-principles/</a>) in the context of big biodata for a company. (for research and advancement as well!)</p><p>His emphasis on a “solid data foundation” requires FAIR data management to be fair. The omics data management repository structure was described in 3 categories: metadata with knowledge graphs, large files with cloud-based systems, and derived results in the data warehouse.</p><p><strong>Single Cell Sequencing of Bacterial Populations</strong></p><p>I participated in this talk just for fun. It turns out that single-cell sequencing is becoming popular for bacteria as well.</p><p>As Matt explained, these are very small cells and mostly ribosomal RNA and no particular poly-A dominated, so requires further fine-tuning of the established protocols. Nevertheless, they accomplished FACS-based single-cell isolation following individually sequencing the single clonal population.</p><p>After Barynard experiments, error-filtering using bulk-seq as a reference, and further confirmations of sequencing went well, they drew the phylogeny of this individual bacteria of a single clone. Surprisingly (or not), they found heterogeneity and could generate the population structure of a single clone. They further confirmed whether cell number is enough to identify variance using hypermutator sampling. It might be helpful to uncover driving small bacterial populations in resistance or certain conditions. (more info: <a href="https://pmc.ncbi.nlm.nih.gov/articles/PMC9676037/">https://pmc.ncbi.nlm.nih.gov/articles/PMC9676037/</a>)</p><p><strong>Advancing Rare Disease Research With Omics Technologies</strong></p><p>I learned about NIHR Biosource for rare disease research (<a href="https://bioresource.nihr.ac.uk/">https://bioresource.nihr.ac.uk</a>), hosting transcriptomics, genomics, epigenomics, proteomics (and metabolomics?) data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*P6UP9cR3_8VQzjsz" /><figcaption>Star Wars themed zone at FoG2025London</figcaption></figure><p>There were genome, single-cell, and spatial-omics-focused domes for improved smaller group interactions and chats. I could not join them as the space was not big, and there was hype on the topics.</p><p>Overall, I enjoyed and learned a lot. I would like to thank everyone for making this event possible. #FoG2025 #FoG2025London</p><p>The most disappointing part could be the visa requirement to be able to join this conference in UK. I know a lot of brilliant colleagues who wanted to attend. However, they could not join due to visa and it’s fees. I hope there are also more solution-oritented policies on FAIR conference opportunities (not just data management and sharing).</p><p>I am looking forward to attending the next exciting conference!</p><p><em>Fun fact: This was my first time attending the Festival of Genomics (FoG). I realized that I was the only researcher(academic) from Cork(Ireland).</em></p><p>BONUS: Brian Cox was the final speaker. I did not know there were two different approaches to space exploration by super rich. You can google more how Musk versus Bezos approach this.. The discussion about “what is time” was surely fun. However, the discussion about the possibility of life in different formats as an example about chirality of nucleic acids, the foundation of the discussion, was a bit misleading. I am not the expert, but technically they exists in either one of the formats dominating (right or left) in different contexts/nucleic acids (e.g., only left for one type of nucleic acid and only right for the other but not always right or left for all, and possibly related to magnetism). If you have ever read any of Dr. Ozturk (Harvard Kavli Prize winner) papers about the topic, you might have a better idea.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1e18f7f8bd26" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Circular RNA Translation]]></title>
            <link>https://fatmab-dincaslan.medium.com/circular-rna-translation-b1d671019664?source=rss-d035227fb6ad------2</link>
            <guid isPermaLink="false">https://medium.com/p/b1d671019664</guid>
            <category><![CDATA[circular-rna]]></category>
            <category><![CDATA[therapeutics]]></category>
            <category><![CDATA[translation]]></category>
            <dc:creator><![CDATA[Ortaya Karışık (Fatma Betul Dincaslan)]]></dc:creator>
            <pubDate>Mon, 13 Jan 2025 19:06:01 GMT</pubDate>
            <atom:updated>2025-01-13T19:06:01.164Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>I published it <a href="https://www.linkedin.com/pulse/circular-rna-translation-fatma-betul-dincaslan-phd--ss44c/?trackingId=3THo59HkT2Wbez35MGB9iQ%3D%3D">here</a>. There has always been some recent news circulating on social media regarding different RNA biotypes. I am lucky to meet incredible researchers from my current lab at UCC to discuss something exciting every time. Moreover, I have been reading about circular RNAs (circRNA) for a while. I would like to share my understanding of the “translation of circular RNA”.</blockquote><p><em>Fasten your belt because we need to learn some basics before diving in! There are several aspects. I will not go into the details of circRNA biogenesis.</em></p><p>Before describing the typical/well-known/canonical or less-known/non-canonical modes of translation, let’s understand why translation matters first. If you are familiar with the central dogma, DNA*-&gt;(transcription)RNA-&gt;(translation)protein, proteins are the critical functional units/doers of a cell.</p><p><em>Fun fact: Although there are various short and long RNA biotypes, we usually refer to mRNA when we talk about coding, and the rest are known as non-coding biotypes. The part of mRNA where it is translated into protein is called coding sequences/region (CDS).</em></p><h4>Let&#39;s go back to the existential question of this post! <br>Does circular RNA even translate? <br>If yes, how do we know?</h4><h3>Modes of Translation in Eukaryotes</h3><blockquote>The best-known cap-independent translation initiation mechanisms are promoted by Internal Ribosome Entry Sites (also known as IRES) through <br>either IRES binding auxiliary trans-acting sites (ITAFs) or epigenetic marks such as N6-methyladenosine (m6A) modification (which then interacts with/recruits some other translation initiation factors) at potential upstream open reading frames (uORF) within 5&#39; Untranslated Region (UTR).</blockquote><p>There are typical and less-known ways of how to read (or translate) the information shared in RNA: cap-dependent and cap-independent translation (CIT).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Tq8GpFIafoU3aN4Z" /><figcaption>mRNA has methyl guanosine cap for 5&#39; (prime) and polyA tail modification for 3&#39; (prime) to be loaded by initiation and recruiting factors bringing two ends closer &gt; processed by pre-initiation complex (PIC) and translate the information encoded</figcaption></figure><p>There are two crucial components of mRNA for the stability and processing during cap-dependent translation of mRNA:</p><ul><li>5&#39;methyl guanosine (m7G) cap</li><li>poly-A tail</li></ul><p>However, it turns out that there are alternative mechanisms present for RNA translation, especially when the canonical translation is somehow blocked (i.e., cap initiation factors are not able to be recruited or shut down due to viral infection) or less efficient. These can be categorized as non-canonical ways of translation. There are still lots of unknowns about them.</p><p>The best-known cap-independent translation initiation mechanisms are promoted by Internal Ribosome Entry Sites (also known as IRES) through <br>either IRES binding auxiliary trans-acting sites (ITAFs) or epigenetic marks such as N6-methyladenosine (m6A) modification (which then interacts with/recruits some other translation initiation factors) at potential upstream open reading frames (uORF) within 5&#39; Untranslated Region (UTR).</p><p>Most of the translation relies on cap-dependent translation. The cap-independent translation usually takes part in certain abnormal conditions such as starvation, oxidative stress, and heat shock.</p><p>This is where circRNAs become interesting.</p><p><em>Fun fact: As endogenous IRES might be rare, some researchers proposed short IRES-like elements sufficient to drive the cap-independent translation.</em></p><h3>Why Does circRNA Matter</h3><blockquote>Yet, the recent advancements in sequencing, data analysis, experimental techniques such as ribosome profiling, and studies on the potential of cap-independent translation showed that some circular RNAs are coding proteins and translated.</blockquote><p>As the name suggests, some RNA has a single-stranded unique circular structure and is expressed endogenously, naturally found in cells. Since its 5&#39; and 3&#39; ends are covalently linked as a loop and lack 5&#39; or 3&#39; free ends, circRNA has shown to have better stability compared to mRNA counterparts (e.g., against exonucleases). However, circRNA was proposed as a non-coding RNA biotype with no known protein-coding function until the last decade.</p><p>Yet, the recent advancements in sequencing, data analysis, experimental techniques such as ribosome profiling, and studies on the potential of cap-independent translation showed that some circular RNAs are coding proteins and translated.</p><p>As circular closed-loop RNA must rely on cap-independent translation (CIT), it will able to be easily detected/used in conditions stimulating CIT.</p><p><em>Fun fact: RNA can also be circularized synthetically as a therapeutic tool with enzymatic ligation or self-splicing.</em></p><h3>How Do We Know circRNA Translates?</h3><p>There are several good practices have been developed over the years regarding how to predict or prove the translation of natural cirRNAs:</p><ol><li>ribosome/polysome binding/occupancy</li><li>experimentally mapped translation initiation sites (TIS)</li><li>(potential) internal ribosome entry site (IRES)</li><li>(potential) N-6-methyladenosine modification sites (m6A) to promote translation initiation</li><li>(potential) open reading frames (ORF)</li><li>mass spectrometry (MS) detecting the circRNA encoded peptides across back-splice junctions (BSJ)</li></ol><p><em>Fun fact: You might be wondering how people can make sure whether they detect any coding linear RNA or circRNA. First, exonuclease treatment and several other approaches are used to digest linear RNA/enrich circRNA during the isolation process. Secondly, most endogenous circRNA are back-spliced(BS) from linear counterparts, so total-RNA seq data with detecting BS events or long-read sequencing is useful.</em></p><p>The last time I took an up-to-date molecular biology course was ~10 years ago when textbook information was not updated on certain topics. I think the field is advancing quite fast. I enjoyed refreshing my knowledge of the basics and application of our current understanding of RNA therapeutics. I hope you have fun reading it, too!</p><h3>References/Further Reading:</h3><ul><li>Why does translatable circular RNA matter? :</li></ul><p>The therapeutic potential of circular RNAs, <a href="https://www.nature.com/articles/s41576-024-00806-x#:~:text=circRNAs%20have%20great%20therapeutic%20potential,candidates%20for%20sustained%20therapeutic%20interventions">https://www.nature.com/articles/s41576-024-00806-x</a></p><ul><li>Central dogma, nucleic acid, protein, translation (briefly): <a href="https://www.genome.gov/genetics-glossary/Central-Dogma#:~:text=The%20central%20dogma%20of%20molecular,National%20Human%20Genome%20Research%20Institute">https://www.genome.gov/genetics-glossary/Central-Dogma#:~:text=The%20central%20dogma%20of%20molecular,National%20Human%20Genome%20Research%20Institute</a></li><li>My previous post about the non-canonical way of translation summarized an article: <a href="https://www.linkedin.com/posts/fbdincaslan_the-dark-proteome-translation-from-noncanonical-activity-7271979603877871616-KGiU">https://www.linkedin.com/posts/fbdincaslan_the-dark-proteome-translation-from-noncanonical-activity-7271979603877871616-KGiU</a>?</li><li>A nice YouTube tutorial video about Cap-Independent Translation Initiation — IRES (virus and eukaryote) and uORF, <a href="https://www.youtube.com/watch?v=3nhOGTLpEi4">https://www.youtube.com/watch?v=3nhOGTLpEi4</a></li><li>If you ever wonder why you don’t capture circular RNA translation in regular conditions that much (a few examples with relevant discussions):</li></ul><p>CircRNA-protein complexes: IMP3 protein component defines subfamily of circRNPs, <a href="https://www.nature.com/articles/srep31313">https://www.nature.com/articles/srep31313</a></p><p>Hypoxia-induced circRNF13 promotes the progression and glycolysis of pancreatic cancer, <a href="https://www.nature.com/articles/s12276-022-00877-y">https://www.nature.com/articles/s12276-022-00877-y</a></p><p>A novel protein encoded by the circular form of the SHPRH gene suppresses glioma tumorigenesis, <a href="https://www.nature.com/articles/s41388-017-0019-9#Sec2">https://www.nature.com/articles/s41388-017-0019-9#Sec2</a></p><ul><li>Translation of CircRNAs, <a href="https://www.sciencedirect.com/science/article/pii/S1097276517301363?via%3Dihub">https://www.sciencedirect.com/science/article/pii/S1097276517301363?via%3Dihub</a></li><li>Pervasive translation of circular RNAs driven by short IRES-like elements, <a href="https://www.nature.com/articles/s41467-022-31327-y#author-information">https://www.nature.com/articles/s41467-022-31327-y</a></li><li>Where RNA modifications, specifically the methylation, join the chat:</li></ul><p>Extensive translation of circular RNAs driven by N6-methyladenosine, <a href="https://www.nature.com/articles/cr201731">https://www.nature.com/articles/cr201731</a></p><p>Circ-ZNF609 Is a Circular RNA that Can Be Translated and Functions in Myogenesis, <a href="https://www.sciencedirect.com/science/article/pii/S1097276517301326?via%3Dihub">https://www.sciencedirect.com/science/article/pii/S1097276517301326?via%3Dihub</a></p><ul><li>A peptide encoded by circular form of LINC-PINT suppresses oncogenic transcriptional elongation in glioblastoma, <a href="https://www.nature.com/articles/s41467-018-06862-2#author-information">https://www.nature.com/articles/s41467-018-06862-2</a></li><li>Structured elements drive extensive circular RNA translation, <a href="https://www.sciencedirect.com/science/article/pii/S1097276521006262">https://www.sciencedirect.com/science/article/pii/S1097276521006262</a></li><li>Fantastic circRNAs and Where to Find Them: Some circRNA identification tools</li></ul><p>CircPrimer 2.0: a software for annotating circRNAs and predicting translation potential of circRNAs, <a href="https://bmcbioinformatics.biomedcentral.com/articles/10.1186/s12859-022-04705-y">https://bmcbioinformatics.biomedcentral.com/articles/10.1186/s12859-022-04705-y</a></p><p>Full-length circular RNA profiling by nanopore sequencing with CIRI-long, <a href="https://www.nature.com/articles/s41596-023-00815-w">https://www.nature.com/articles/s41596-023-00815-w</a></p><p>Best practice standards for circular RNA research, <a href="https://www.nature.com/articles/s41592-022-01487-2#Tab2">https://www.nature.com/articles/s41592-022-01487-2#Tab2</a></p><p>An updated resource for the detection of protein-coding circRNA with CircProPlus, <a href="https://www.nature.com/articles/s41598-024-69744-2">https://www.nature.com/articles/s41598-024-69744-2</a></p><p>circFL-seq reveals full-length circular RNAs with rolling circular reverse transcription and nanopore sequencing, <a href="https://elifesciences.org/articles/69457">https://elifesciences.org/articles/69457</a></p><p>TransCirc: an interactive database for translatable circular RNAs based on multi-omics evidence, <a href="https://academic.oup.com/nar/article/49/D1/D236/5930392?login=true">https://academic.oup.com/nar/article/49/D1/D236/5930392?login=true</a> (I liked the database idea so much. However, they lack links to refer to evidence where claimed, which article, etc.)</p><ul><li>The molecular basis of translation initiation and its regulation in eukaryotes, <a href="https://www.nature.com/articles/s41580-023-00624-9">https://www.nature.com/articles/s41580-023-00624-9</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b1d671019664" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Single Cell Experiments: Spike-in]]></title>
            <link>https://fatmab-dincaslan.medium.com/single-cell-experiments-spike-in-7998fdf3edff?source=rss-d035227fb6ad------2</link>
            <guid isPermaLink="false">https://medium.com/p/7998fdf3edff</guid>
            <category><![CDATA[quality-control]]></category>
            <category><![CDATA[spike-in]]></category>
            <category><![CDATA[sequencing]]></category>
            <category><![CDATA[rna-sequencing]]></category>
            <category><![CDATA[library-preparation]]></category>
            <dc:creator><![CDATA[Ortaya Karışık (Fatma Betul Dincaslan)]]></dc:creator>
            <pubDate>Thu, 26 Dec 2024 17:41:25 GMT</pubDate>
            <atom:updated>2024-12-26T17:41:25.036Z</atom:updated>
            <content:encoded><![CDATA[<p>I shared it on <a href="https://www.linkedin.com/pulse/single-cell-experiments-spike-in-fatma-betul-dincaslan-phd--fovxc/?trackingId=6%2BhmXHucR1mgQg7JiIYNcQ%3D%3D">my LinkedIn page </a>first. <em>Although there are many protocols and guidelines on sequencing, if you are not preparing the sample for the sequencing machine to set up, you might not know what spike-in is. No worries, I am here to explain this. I need to learn these details (including troubleshooting the sequencing itself, and still learning) because my PhD experience covers molecular assay development and low-bulk optimization to sample preparation for sequencing (including maintaining the Miseq device (Illumina)).</em></p><p>Spike-ins are external nucleic acids. Depending on the purpose, they can be <strong>RNA- or DNA-based</strong>. They are used mainly for <strong>normalization</strong>, <strong>quantification</strong>, and <strong>sequencing calibration/control</strong>.</p><p><em>Let’s learn the details of the commonly used spike-ins in transcriptomics sequencing: </em><strong>ERCC</strong> and <strong>PhiX</strong><em>.</em></p><p><strong>ERCC</strong></p><p>Since internal controls such as housekeeping genes might vary expression and primer affinities might differ, researchers came up with the idea of external controls. The External RNA Controls Consortium (ERCC) is an RNA spike-in with <strong>known sequence </strong>(with minimal sequence homology with endogenous transcripts of most of the popular organisms)<strong>, sense strand, and concentrations</strong>. It is <strong>added to RNA </strong>samples as a reference (while they were still RNA, not converted to cDNA, etc.) to track the variation. They are <strong>poly-A-tailed</strong> to be captured just like mRNA.</p><p>Where to add ERCC spike-in</p><p>It is particularly useful for <strong>detecting the sensitivity </strong>(e.g., for low abundance/limited number transcripts), <strong>eliminating the technical variability across samples and platforms </strong>due to initial sample input or sequencing depth, <strong>linearity of amplifications</strong>, <strong>strandedness accuracy of alignments</strong>, and <strong>absolute quantification</strong>. It might not only be used in Next Generation Sequencing (NGS) but also some microarrays and RT-PCRs.</p><p>You can use custom-designed RNA spike-ins as an alternative.</p><p><em>Do we always need ERCC as a control</em>? Not exactly. In most cases of RNA-seq data differential gene expression analysis, we rely on relative gene abundance for the comparison. We use sequencing depth normalizations across samples in most cases. If you need absolute gene quantification, then yes. Moreover, you need to consider bias can be introduced due to ERCC spike-in itself (e.g., due to manual addition difference, low diversity, an additional layer of complexity, and the requirement for the calibration for varying abundance).</p><p><strong>PhiX</strong></p><p>PhiX is a bacteriophage with single-stranded DNA. It has a balanced base diversity (~45% G/C and ~55% A/T). It is used to check the overall quality of Illumina sequencing runs itself (It has nothing to do with sample preparation and library normalization).</p><p>Where to add PhiX spike-in</p><p>Sometimes, the library you are preparing is low diversity. For example, I was trying to figure out total RNA sequencing in various formats, one of them was using miRNA with external polyA extension from another species (e.g., cel-mir-37). If you try to sequence such a sample with a small number of cDNA-based library or imbalanced base concentrations (A/T/C/G) due to unique library preparation, Illumina machines fail. You might need to add higher concentrations of <strong>PhiX to increase the diversity</strong> and <strong>balanced nucleotide ratio </strong>up to 50%. Even if you have enough diversity in the library, you might still need to <strong>monitor the quality of sequencing</strong>, <strong>base calling </strong>(e.g., Q30 scores), and <strong>estimate some sequencing by synthesis (SBS) metrics</strong> in this case, as low as 1% would be enough.</p><p><em>Fun fact: PhiX was the first DNA-based genome to be sequenced by Sanger.</em></p><p>In summary:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZJ4Fz_1ZO3slY3U7HO6fRw.png" /><figcaption>Although we start with RNA, the final library we sequence is converted to DNA for Illumina sequencing. ERCC is added while the sample is at the initial RNA stage, whereas PhiX is added at the final stage for the sequencing run.</figcaption></figure><p><strong>Further Reading/ References:</strong></p><ul><li>ERCC, <a href="https://www.nist.gov/programs-projects/external-rna-controls-consortium"><strong>https://www.nist.gov/programs-projects/external-rna-controls-consortium</strong></a> and the initial release: <a href="https://bmcgenomics.biomedcentral.com/articles/10.1186/1471-2164-6-150"><strong>https://bmcgenomics.biomedcentral.com/articles/10.1186/1471-2164-6-150</strong></a></li><li>ERCC Thermo Fisher detailed manual, <a href="https://assets.thermofisher.com/TFS-Assets/LSG/manuals/cms_086340.pdf"><strong>https://assets.thermofisher.com/TFS-Assets/LSG/manuals/cms_086340.pdf</strong></a> , and Lexogen detailed sequencing manual with spike-in details: <a href="https://www.lexogen.com/wp-content/uploads/2020/01/113UG227V0100_QuantSeq-FWD_UDI_Kits_2019-12-27.pdf"><strong>https://www.lexogen.com/wp-content/uploads/2020/01/113UG227V0100_QuantSeq-FWD_UDI_Kits_2019-12-27.pdf</strong></a></li><li>Where and how to use ERCC (briefly), <a href="https://genomics.rwth-aachen.de/site/ercc/"><strong>https://genomics.rwth-aachen.de/site/ercc/</strong></a>, <a href="https://pmc.ncbi.nlm.nih.gov/articles/PMC4760223/#sec7"><strong>https://pmc.ncbi.nlm.nih.gov/articles/PMC4760223/#sec7</strong></a></li><li>PhiX, <a href="https://en.wikipedia.org/wiki/Phi_X_174"><strong>https://en.wikipedia.org/wiki/Phi_X_174</strong></a>, and more about its sequencing applications: <a href="https://knowledge.illumina.com/library-preparation/general/library-preparation-general-reference_material-list/000001545"><strong>https://knowledge.illumina.com/library-preparation/general/library-preparation-general-reference_material-list/000001545</strong></a>, <a href="https://www.illumina.com/products/by-type/sequencing-kits/cluster-gen-sequencing-reagents/phix-control-v3.html"><strong>https://www.illumina.com/products/by-type/sequencing-kits/cluster-gen-sequencing-reagents/phix-control-v3.html</strong></a></li><li>Where to be careful about PhiX, <a href="https://environmentalmicrobiome.biomedcentral.com/articles/10.1186/1944-3277-10-18#:~:text=Due%20to%20its%20small%2C%20well,40%25%20for%20low%20diversity%20samples"><strong>https://environmentalmicrobiome.biomedcentral.com/articles/10.1186/1944-3277-10-18#:~:text=Due%20to%20its%20small%2C%20well,40%25%20for%20low%20diversity%20samples</strong></a>.</li><li>Bonus: Unique Molecular Identifiers (UMI) to be spiked-in: <a href="https://www.nature.com/articles/s41592-022-01446-x"><strong>https://www.nature.com/articles/s41592-022-01446-x</strong></a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7998fdf3edff" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why and How mRNA-Protein Ratio Matter]]></title>
            <link>https://fatmab-dincaslan.medium.com/why-and-how-mrna-protein-abundance-matter-351fdc5b0474?source=rss-d035227fb6ad------2</link>
            <guid isPermaLink="false">https://medium.com/p/351fdc5b0474</guid>
            <category><![CDATA[protein]]></category>
            <category><![CDATA[omics]]></category>
            <category><![CDATA[mrna]]></category>
            <category><![CDATA[omics-data-analysis]]></category>
            <category><![CDATA[expression]]></category>
            <dc:creator><![CDATA[Ortaya Karışık (Fatma Betul Dincaslan)]]></dc:creator>
            <pubDate>Thu, 14 Nov 2024 21:55:17 GMT</pubDate>
            <atom:updated>2024-11-14T22:36:36.011Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>First shared <a href="https://www.linkedin.com/pulse/why-how-mrna-protein-correlation-matter-fatma-betul-dincaslan-phd--ppk0c/?trackingId=bmIMQpQARSmqT9XNY4%2Bxtw%3D%3D">here</a>. I have been planning several posts, yet not ready for any of them. I decided to use one of AI tools, Write For Me, for the first time to write this blog post. I gave the relevant links and bullet points to the ChatGPT prompt, then asked for a 1000-word blog post. I need to give feedback and do revisions (and make it shorter later) as well. However, we finally did it. I will share the initial prompt, just in case you are wondering. Let’s see how it worked!</blockquote><p>The accurate interpretation of mRNA and protein levels is a critical challenge in modern biology. While mRNA serves as a blueprint for protein synthesis, the correlation between mRNA and protein abundance is only moderate, complicating the design of translation-dependent experiments. Insights from multiple studies show that relying solely on mRNA data can lead to misleading conclusions, particularly in complex biological contexts like tissue specificity and disease research.</p><h3>Why mRNA-Protein Correlation Is Limited</h3><p>mRNA is often used to estimate protein levels, but studies show that this relationship is far from perfect. The analysis of 32 tissues revealed a correlation of 0.46 between mRNA and protein, suggesting that mRNA explains only a part of protein abundance. Measurement accuracy significantly influences the observed correlations. For example, in tumor proteomics, highly reproducible datasets showed better mRNA-protein alignment, emphasizing the need for rigorous validation of omics data.</p><p>Biological factors also complicate the relationship. Protein turnover and regulatory mechanisms such as translation initiation, mRNA stability, and post-translational modifications all contribute to discrepancies between mRNA and protein levels. Tissue/condition-specific phosphorylation further adds complexity, enabling proteins to adapt to distinct cellular environments independent of mRNA.</p><h3>Key Factors Affecting mRNA-Protein Disparities</h3><h3>mRNA and Protein Turnover</h3><p>Differences in the degradation and synthesis rates of mRNA and proteins can lead to inconsistent levels. While some proteins have long half-lives, their mRNA may be rapidly degraded, and vice versa. This creates a disconnect that highlights the limitations of using mRNA as a sole proxy for protein levels.</p><h3>Tissue and Condition Specificity</h3><p>Gene expression varies significantly between tissues and under different conditions. Proteins may be more abundant in certain tissues due to context-specific translation, while their mRNA levels remain stable. A study found that tissue-specific phosphorylation allows for fine-tuning of protein function, influencing the interpretation of gene expression. For instance, phosphorylated proteins tend to be expressed more ubiquitously, while non-phosphorylated forms are often tissue-specific, fine-tuning the cell’s performance to local needs. This complexity underscores the importance of integrating proteomic data with transcriptomic data for a comprehensive understanding.</p><h3>Translation Efficiency and Post-Translational Modifications</h3><p>Regulatory elements like microRNAs and RNA-binding proteins can affect how efficiently mRNA is translated into proteins. Additionally, modifications like phosphorylation alter protein function, adding complexity to the interpretation of gene expression data.</p><h3>Protein Mass and Cell Size</h3><p>Larger cells generally contain more protein mass, which impacts absolute protein levels. While this does not directly affect the mRNA-protein ratio, it can influence the interpretation of experimental data, particularly when comparing across different cell types. In some cases, histone levels are used as a normalization factor because they correlate with DNA content rather than cell size, offering a more stable baseline for comparison (also known as “proteomics ruler)”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*CtFlNztqoq6NNnhL" /><figcaption>Understanding the disparities</figcaption></figure><h3>Improving Experimental Design and Data Interpretation</h3><p>To overcome the challenges of mRNA-protein disparities, researchers are increasingly turning to improved reproducibility, intra-sample instead of inter-sample RNA-protein correlations, considering context specificity and integrated omics approaches.</p><p>Visualization tools like GeneRanger provide context-specific insights, allowing scientists to assess the translation potential of genes in specific tissues. The Human Protein Atlas and other specialized databases offer comprehensive datasets that enhance the reliability of gene expression analysis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*-LSEKNNMzr--8y7W" /><figcaption>Although Gene Ranger covers HPA as well, I enjoy HPA interface for exploring TS. Tabula Sapiens dataviz is also a good alternative. SPIDERP package can also be used for constructing cell-type-specific protein-protein interaction networks.</figcaption></figure><h3>Conclusion</h3><p>The mRNA-protein relationship is complex, with numerous factors affecting protein abundance beyond transcription. For accurate experimental design, especially in functional genomics and disease research, integrating transcriptomic and proteomic data is essential. By utilizing context-specific datasets and considering the biological and technical factors that influence gene expression, scientists can make more reliable predictions and better understand cellular behavior.</p><p>As you see below, I have already made a good selection of resources and summaries for it to organize the content. So, I am not sure whether it saved my time or not.</p><h3>References/ Prompt from my Summaries and Selections</h3><p>Protein vs mRNA levels, why need to be careful while translation-dependent experiment designs. Why I am obsessed with mRNA vs protein level disparities:</p><ul><li>The utility of protein and mRNA correlation, <a href="https://pdf.sciencedirectassets.com/271294/1-s2.0-S0968000414X00137/1-s2.0-S09680004140020[%E2%80%A6]uY29t&amp;ua=13035903555703095005&amp;rr=8e16af2afed09552&amp;cc=ie">https://pdf.sciencedirectassets.com/271294/1-s2.0-S0968000414X00137/1-s2.0-S09680004140020[…]uY29t&amp;ua=13035903555703095005&amp;rr=8e16af2afed09552&amp;cc=ie</a></li><li>Insights into the regulation of protein abundance from proteomic and transcriptomic analyses, <a href="https://www.nature.com/articles/nrg3185.pdf?utm_source=sciencedirect_contenthosting&amp;getft_integrator=sciencedirect_contenthosting">https://www.nature.com/articles/nrg3185.pdf?utm_source=sciencedirect_contenthosting&amp;getft_integrator=sciencedirect_contenthosting</a></li><li>A deep proteome and transcriptome abundance atlas of 29 healthy human tissues, <a href="https://www.embopress.org/doi/full/10.15252/msb.20188503">https://www.embopress.org/doi/full/10.15252/msb.20188503</a></li><li>A Quantitative Proteome Map of the Human Body, <a href="https://www.sciencedirect.com/science/article/pii/S0092867420310783?via%3Dihub#bib57">https://www.sciencedirect.com/science/article/pii/S0092867420310783?via%3Dihub#bib57</a> “We computed the correlation between the protein and RNA abundance across 32 tissues for each gene and found the median Spearman correlation is 0.46 (interquartile range of 0.24–0.65), consistent with previous findings “</li><li>An objection: “Experimental reproducibility limits the correlation between mRNA and protein abundances in tumor proteomic profiles”, <a href="https://www.sciencedirect.com/science/article/pii/S2667237522001709?via%3Dihub">https://www.sciencedirect.com/science/article/pii/S2667237522001709?via%3Dihub</a> “more reproducibly measured proteins have higher mRNA-protein correlation, suggesting that measurement error limits mRNA-protein correlation”</li><li>If you are looking for protein-level (translation potential) tissue/cell-specific databases (apart from the human proteome atlas), these guys help you to visualize the selected gene/protein level at various protein and transcriptomics level databases: <a href="https://generanger.maayanlab.cloud/gene/TEK?database=GTEx_proteomics">https://generanger.maayanlab.cloud/gene/TEK?database=GTEx_proteomics</a>, <a href="https://academic.oup.com/nar/article/51/W1/W213/7160193">https://academic.oup.com/nar/article/51/W1/W213/7160193</a></li><li>SPIDER: constructing cell-type-specific protein–protein interaction networks, <a href="https://academic.oup.com/bioinformaticsadvances/article/4/1/vbae130/7746021">https://academic.oup.com/bioinformaticsadvances/article/4/1/vbae130/7746021</a></li></ul><p>How to correlate/interpret protein-mRNA levels?</p><ul><li>Protein mass changes with the cell size, tissue, and condition</li><li>mRNA and protein turnover rates are also important</li><li>The histone levels are sometimes used to have a better comparison of the profiles because histone copy number is independent of cell size (proportional to DNA amount rather than the volume)</li></ul><p>Condition or disease specificity to be discussed as well</p><ul><li>Signaling plasticity in the integrated stress response, <a href="https://www.frontiersin.org/journals/cell-and-developmental-biology/articles/10.3389/fcell.2023.1271141/full#B80">https://www.frontiersin.org/journals/cell-and-developmental-biology/articles/10.3389/fcell.2023.1271141/full#B80</a></li></ul><p>Expression Atlas, some proteomics comprehensive datasets</p><ul><li>A tissue-specific atlas of mouse protein phosphorylation and expression, <a href="https://www.ebi.ac.uk/gxa/experiments/E-PROT-13/Results">https://www.ebi.ac.uk/gxa/experiments/E-PROT-13/Results</a>, <a href="https://europepmc.org/article/MED/21183079">https://europepmc.org/article/MED/21183079</a> i) Half of all sites are unique to a single tissue, and few sites are globally phosphorylated. ii) Classes of sites are differentially enriched in different tissues. iii) Distinct kinases and phosphatases are present in each tissue. iv) Both individual proteins and entire protein complexes show tissue-specific phosphorylation. Phosphorylated and non-phosphorylated proteins display markedly different expression patterns. Phosphoproteins are more often expressed globally, suggesting that tissue-specific phosphorylation allows tuning of ubiquitous proteins to optimize cell performance. Together, complementary protein expression and phosphorylation maintain the unique properties of distinct tissues.</li><li>Human immunochemistry data on 83 different normal cell types from 44 tissue types from the Human Protein Atlas project, <a href="https://www.ebi.ac.uk/gxa/experiments/E-PROT-3/Results">https://www.ebi.ac.uk/gxa/experiments/E-PROT-3/Results</a>, <a href="https://www.nature.com/articles/nature13302">https://www.nature.com/articles/nature13302</a> and <a href="https://www.science.org/doi/10.1126/science.1260419">https://www.science.org/doi/10.1126/science.1260419</a></li><li>Individual variability of protein expression in human tissues, <a href="https://www.ebi.ac.uk/gxa/experiments/E-PROT-43/Results">https://www.ebi.ac.uk/gxa/experiments/E-PROT-43/Results</a>, <a href="https://pubs.acs.org/doi/10.1021/acs.jproteome.8b00580">https://pubs.acs.org/doi/10.1021/acs.jproteome.8b00580</a> (only 9 tissues)</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=351fdc5b0474" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>