<?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 Live Data Concepts on Medium]]></title>
        <description><![CDATA[Stories by Live Data Concepts on Medium]]></description>
        <link>https://medium.com/@LiveDataConcept?source=rss-7b56a566de9f------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*QY96htiZ8c-wUVxh.png</url>
            <title>Stories by Live Data Concepts on Medium</title>
            <link>https://medium.com/@LiveDataConcept?source=rss-7b56a566de9f------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 22:31:55 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@LiveDataConcept/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[Drug Development Knowledge Graph insights using visual and interactive cues]]></title>
            <link>https://medium.com/@LiveDataConcept/drug-development-knowledge-graph-insights-using-visual-and-interactive-cues-df1707b5e039?source=rss-7b56a566de9f------2</link>
            <guid isPermaLink="false">https://medium.com/p/df1707b5e039</guid>
            <category><![CDATA[competitive-intelligence]]></category>
            <category><![CDATA[visualization]]></category>
            <category><![CDATA[knowledge-graph]]></category>
            <category><![CDATA[drug-development]]></category>
            <category><![CDATA[neo4j]]></category>
            <dc:creator><![CDATA[Live Data Concepts]]></dc:creator>
            <pubDate>Tue, 31 Mar 2020 18:00:39 GMT</pubDate>
            <atom:updated>2020-03-31T18:00:39.974Z</atom:updated>
            <content:encoded><![CDATA[<p>So, you have a knowledge graph, now what are you going to do with it? A knowledge graph, a collection of interconnected concepts representing a particular knowledge domain, should be able to give you insights into what you are looking for and more. You ought to be able to look up concepts, gather what is known about the individual “thing” and be able to navigate to see connections to other things. Most example graph visualizations use rather simplified scenarios to show immediate connections whereas in the real world one typically deals with increasingly complex pictures, especially as knowledge capture within the graph increases. One then requires having a visualization tool which enables application of visual cues to each concept in a graph in order to capture more knowledge within a node and relationship representation itself. Furthermore, the tool must enable the user to navigate the connected landscape interactively via the application of searches and filters that eliminate unnecessary or undesired information within the graph visualization.</p><p>In a <a href="https://medium.com/@LiveDataConcept/visual-interrogation-of-a-cancer-knowledge-graph-using-neo4j-bloom-d00687969da2">previous blog post</a> I showed some examples from a competitive intelligence knowledge graph covering the biotech/pharma industry focusing on cancer drugs in development. I used Neo4j Bloom v1.0 to determine how well one can navigate the content to gain insights. However, I also learned that while simple scenarios can be handled very well, when it comes to navigating beyond a one-path relationship neighborhood, one rapidly encounters that typical “hairball” problem which leads to uninterpretable graphs or very laborious workarounds to clean up things not needed.</p><p>Here I am following up on this story to showcase new features in Neo4j Bloom v1.2.1 which now make an interactive graph exploration so much more insightful by 1) enabling the addition of visual cues on nodes and relationships and 2) providing new “Expand” interactivity to grow a graph from <strong>multiple selected nodes</strong> based on <strong>one or multiple specific relationship types</strong>. These new features are a major step towards avoiding the typical “graph hairball” problems and being able to interactively leverage the full power of a knowledge graph.</p><h3><strong>A brief introduction to the graph data model</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Hm60eKHR3iBKgPYmjxKDrQ.png" /></figure><p>As outlined before at the heart of this knowledge graph is the connection between a <strong>Company</strong>, a <strong>Therapeutic Molecule</strong>, a drug the company is developing and a <strong>Molecular Target</strong> the drug molecule is acting on.</p><p>As multiple companies tend to work on molecules that act on the same target one can generate a drug development network.</p><h3><strong>Visual cues using qualitative measures enhance graph insights</strong></h3><p>The below image shows a set of companies working on drugs acting on SIRPA, a molecular target for cancer. This example is utilizing visual cues on nodes and relationships to highlight the type of node, the attributes which define the appearance of the node and the type of relationship connecting two nodes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FEKZb3FV6OAO3yC2o6Fs3w.png" /></figure><p>The details can be described to the user in the form of a legend accompanying the image, such as follows:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nRqJRgbyWPuEhg2PSKXDAg.png" /></figure><p>William Lyon describes how to get this done in his Graph Data Science blog, see References below.</p><h3><strong>Predefined searches enable quick graph insights</strong></h3><p>Besides the typical question as to “What do we know about x?” giving insight about a single node, a knowledge graph should also enable views into more complex scenarios. Here we show an example highlighting a common scenario within the biotech/biopharma industry using the above described qualitative measures to style an auto-generated graph answering the question: “What is the current landscape of companies working on a given molecular target?”. We would like to easily see which companies are PRIVATE or PUBLIC and what is the highest Clinical Trial status for each Therapeutic Molecule. Using the case of SIRPA molecular target one can predefine a query that collects not just the target but also the drug molecules acting on the target as well as the companies working on developing the drugs. Using the visual cues as described above one can quickly gain insight into the target landscape in a single picture.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uP_e1DkLvbffgqAFpUm3qA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/714/1*fh68YKBk6tytuK35WuIAVA.png" /></figure><p>This is done by defining a custom Cypher search using a unique name such as “TargetLandscape for SIRPA” in this case. Furthermore, the search can also be defined with the option of having the user select the specific target of interest from a list of available targets. As such this approach and similar ones can become immediately applicable to all available targets within the knowledge graph.</p><h3><strong>Key Performance Indicators used as quantitative styling options</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ppSrPegMXlH7V6UgNA_WSQ.png" /></figure><p>While the previous example focuses on qualitative measures to dynamically adjust the visualization one can also use quantitative measures generated from data within the graph to highlight Key Performance Indicators (KPIs). Let us assume that the number of Clinical Trials for a Therapeutic Molecule is a KPI as to how far advanced a molecule is in its development. For this case we would want to create a separate Perspective with the relevant categories included and style the Therapeutic Molecule nodes by color and size according to an available KPI measure range.</p><h3><strong>Interactive exploration of a target landscape</strong></h3><p>Many initial interactive explorations of a graph start with interest in a specific thing, such as a Company, a Therapeutic Molecule or a Molecular Target. Alternatively, it starts with a predefined search. Immediate neighborhoods, one connection away, are typically explored to add things of further interest. In the example shown we start with a Molecular Target, SIRPA, a target which recently received attention as a possible future oncology target on macrophages. We already see the molecules acting on the target in the previous image. Now we want to add the companies for only the top three molecules by using a new interactive feature to expand the graph by a specific set of relationships as shown below. We only want to expand three nodes based on the relationship “developedByCompany” to add the relevant Company nodes into the picture.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*p4rDvQ33N6K2uzM3Wd53NQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*T4oxK_UOs39iBPrtgIltOg.png" /></figure><p>This is an extremely powerful new Expand feature <strong>acting on multiple nodes, multiple relations and/or categories </strong>to interactively explore further knowledge contained in the graph without overwhelming the graph visualization.</p><p>The combination of Perspectives, Searches, Visual Cues and Interactive Selections on Multiple Nodes make Neo4j Bloom an excellent tool for gaining quick insights into Knowledge Graph contents.</p><h3>But, we always want more ways to gain further insights</h3><p>There is always more to be desired. More data lead to more desired features. Hence here is my but shrinking personal Bloom wish list to make this tool even more powerful for a business analyst:</p><p>· Many business intelligence tools enable visual styling using not only color and size but also by adding shape as an additional visual cue</p><p>· Sometimes it would be helpful if nodes could be displayed using an available image as an overlay (e.g.: Company logo or Product image) instead of using simple icons. This could also be tied to a specific rule-based styling.</p><p>· As knowledge graph content grows, not everything can be communicated in a single styled visualization. Different perspectives can already be used today to handle different graph representations; however, one has to switch between perspectives. A powerful new feature would be the option to enable multiple page views within tabs, or going further, even multiple views within the same window.</p><p>· While an automatic layout algorithm arranges the nodes initially, most often a manual re-arrangement is necessary to create a more static picture of a landscape. It would be fantastic if one could save a manually arranged graph visualization for later retrieval.</p><p>· Last, but not least for the case of displaying dense nodes in a more useable form, enabling grouping of nodes into a single “group” node based on label or category would be another excellent addition.</p><p>Lastly, kudos to the developers of Neo4j Bloom for providing an extremely powerful visualization tool to interactively use dynamic graph visualizations for knowledge discovery and insight.</p><h3>References</h3><ul><li><a href="https://message.neo4j.com/qxCR0Y0P1a0Nbx02R0ZY001">Bloom Documentation</a></li><li><a href="https://sandbox.neo4j.com/?usecase=bloom/?ref=email-&amp;mkt_tok=eyJpIjoiTlROa01HVXlNVGd3TVRnMyIsInQiOiJ6XC91YVEyNlN1SUh2aHZIQ0kzNlF1SUFadFhsVmtiU1hydE1aVDZnM3Yyb2p1QWF3QndFQmZycUtpZnlHWFpTVDEycGJDOWwrZVhwb2lWdEEwUTBYaUxJZ1RPY05nQTF0SXk2NUU0T1BiSWNPT3ZVVWt2WVN2aUtTbkdESlh3bUEifQ%3D%3D">Bloom Sandbox environment</a></li><li><a href="https://medium.com/neo4j/hands-on-with-the-neo4j-graph-data-science-sandbox-7b780be5a44f">Hands-On with the Neo4j Graph Data Science Sandbox</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=df1707b5e039" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Visual interrogation of a cancer knowledge graph using Neo4j Bloom]]></title>
            <link>https://medium.com/@LiveDataConcept/visual-interrogation-of-a-cancer-knowledge-graph-using-neo4j-bloom-d00687969da2?source=rss-7b56a566de9f------2</link>
            <guid isPermaLink="false">https://medium.com/p/d00687969da2</guid>
            <category><![CDATA[knowledge]]></category>
            <category><![CDATA[neo4j]]></category>
            <category><![CDATA[visualization]]></category>
            <category><![CDATA[cancer]]></category>
            <category><![CDATA[discovery]]></category>
            <dc:creator><![CDATA[Live Data Concepts]]></dc:creator>
            <pubDate>Sat, 18 May 2019 23:30:45 GMT</pubDate>
            <atom:updated>2019-05-18T23:30:45.180Z</atom:updated>
            <content:encoded><![CDATA[<h3>What is a knowledge graph?</h3><p>Being a trained molecular biologist I always wondered how one could capture knowledge about a particular domain of interest. For myself this domain constitutes the discovery and development of drugs for cancer. Like for most my endeavor started by setting up a list of drugs and the companies that develop them. Soon, however, it became apparent that a simple list in Excel is not going to suffice to capture an understanding of this knowledge domain.</p><p>Having worked in an Information Technology environment for years I knew that a list typically ends up growing into a database of sorts with tables and relationships connecting them. Naturally as my understanding of the domain increased so did the list, which now had become a collection of lists. Clearly a database was needed to take the next step. Having seen many databases getting defined and populated I knew about the problems relational databases faced when a domain is evolving, but cannot be clearly defined at the outset. I wanted to avoid this problem and got fascinated by the emerging idea of defining nodes and connecting them via relationships, a graph database.</p><p>Luckily for me I came across a nascent database vendor called Neo4j and its database product, which as a community version was and still is completely free. How I managed to get my knowledge from lists into a novel graph database can be a topic for another story. This story is about looking at an accumulated knowledge over several years about biotech and pharma companies, their drugs in development and the clinical trials they run to determine safety and efficacy before getting approval to market and sell them.</p><h3>Neo4j Desktop and Bloom</h3><p>Neo4j Desktop is as the name implies a desktop application to manage a graph database locally or remotely. For simplicity sake I run my database locally, but it could easily be deployed into the cloud. Neo4j comes with a browser application that is setup to visualize its data as a graph and this works well for simple graphs and clearly defined questions. For deeper exploration, especially visual and interactive exploration of database content, from a broad perspective as well as looking at details, another tool is needed. Meet Bloom, a plug-in to Neo4j Desktop. Here is a first look into what I learned visually by interacting with my knowledge domain using Bloom.</p><h3>Biotechs and Pharmas focused on Oncology</h3><p>The universe of companies developing drugs for cancer is substantial, my database covers over 1000 companies, even though I don’t consider my database to be a large database, definitely not BigData. However it is highly connected, meaning there are a lot of relationships connecting various types of data. So an attempt to look at all Companies and their immediate connections results in the following, which I refer to as the typical graph hairball problem.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*d-G0g3TKyXy0iKCSFs8u3Q.png" /><figcaption>Fig.1: An artful hairball of nodes and relationships</figcaption></figure><p>Luckily Bloom has tools that allowed me do the necessary cleanup by removing nodes of types I don’t care about at this time. I wanted to limit my visual to Companies and Drugs only, to get a high level picture of connections in this simplified landscape.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hunzEjeVuUwRdr-JZ873og.png" /><figcaption>Fig.2: Established, mid-stage and start-up companies all together</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/844/1*j0zRNt2btP4MmmMOJUQiXg.png" /><figcaption>Fig.3: A close-nit network</figcaption></figure><p>Connections between companies are represented via molecules two companies are working together on. This can be due to an Event, such as a collaboration or as most often is the case via a Clinical Trial. As one would expect the center is made up of companies that represent the household names of biotech and pharma companies, such as Novartis, Merck, Amgen, Celgene and more. They are highly connected since they tend to work together on the same molecules used in clinical trials. The center is surrounded by smaller companies with a few drugs in development making up biotechs with an early drug development pipeline. Even further outside are companies which do not yet have any drugs in their pipeline, but have received early funding to develop candidate molecules. They are waiting to be connected once they reveal the molecules selected to be developed into a drug. Depending on someones interest one can dedicate their focus of interest on any of these knowledge landscape domains.</p><h3>Delving into the details of an early stage company</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/914/1*Al-xNqCQyemptza3iHMUrw.png" /><figcaption>Fig.4: Company details</figcaption></figure><p>Bloom enables a dive into the details for each node and relationship. Sometimes not much is known about a company, but the knowledge of its existence by itself is important to acknowledge. As more information becomes available it can easily be added to the database and connected with the respective company record or records. The flexibility of dealing with information whether or not it is available and the option of dealing with one or many relationships is a big selling point for Neo4j and makes it super easy to extend an existing data model.</p><h3>Understanding a knowledge neighborhood</h3><p>Molecular Targets are what drug molecules typically act on to influence a disease process. As novel targets get discovered companies start to emerge that see an opportunity to develop therapeutic molecules. By setting up a predefined, parameterized search with auto-suggestions in Bloom one can start to interactively investigate the data landscape around a molecular target.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3e0fjUiqV_17Dbd9J3Vpxg.png" /><figcaption>Fig.5: Defining a basic lookup for Molecular Target by Name</figcaption></figure><p>Starting with a single node I can expand and grow the landscape relationship by relationship, which is crucial as we have seen earlier, hairball graphs are an all too common distraction.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wK3WHViF-FH9IsH2RZSJEA.png" /><figcaption>Fig.6: Expanding a target neighborhood</figcaption></figure><p>A few steps of expanding the graph as well as shrinking it enables me to generate a picture of a connected data landscape representing companies working on developing drugs that target a molecule called CD47, I call it the CD47 star system, others might refer to it as the Molecular Target 360 degree view. Bloom’s tools to expand and dismiss nodes are excellent to trim a graph to a desired shape, almost like getting a great haircut.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*I_9-_GrikJ_k7h3yW6SNzA.png" /><figcaption>Fig.7: A more complete picture of target, molecules and companies with added events</figcaption></figure><h3>My wishlist of Bloom features</h3><p>Like with any tool there are usually a few things that could be added to increase its utility even further. As I was navigating this database to use it for further knowledge discovery I came across the following I would love to see available in Bloom:</p><ul><li>Hyperlinks on property panel to enable links to outside resources, not all information is typically in a single place. Getting to it via single click is crucial.</li><li>Ability to expand multiple nodes by a particular relationship or related node types only would allow for fewer expand and dismiss steps</li><li>Launch multiple windows enabling simultaneous multiple perspectives of the same data. Sometimes different perspectives are needed to fully understand a data landscape, like a browser has many tabs.</li><li>Panning on a layout to view items that disappeared from view</li><li>Availability of additional layout algorithms to change the way the graph layout is organized</li><li>Enable picture overlay on nodes based on picture url. An icon is great to identify what kind of node is represented, but in some cases a logo is preferred.</li></ul><h3>Closing thoughts</h3><p>Connectivity is the name of the game with graphs. In Biology in particular, most “things” in the end will feature a many-to-many relationship between each other even if they don’t start out that way. Schema-based data models, such as the ones developed with relational databases, cannot easily be modified after they have been defined and populated. Graph-based models can evolve easily as more knowledge accumulates. Graph-based, interactive explorations work well to generate 360 views of “things”, such as a Company, Molecular Target, Therapeutic Molecule, Event and so forth, each forming a knowledge star system. Once we connect these star systems and view them from a distance we view a galaxy of knowledge eventually making a knowledge universe. Neo4j Bloom is a lens that lets us view and ask about “things” in this universe from a distance or close up.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d00687969da2" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>