<?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 Andrey Matyash on Medium]]></title>
        <description><![CDATA[Stories by Andrey Matyash on Medium]]></description>
        <link>https://medium.com/@andrej-matyash?source=rss-dc35fb99fab2------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*j54b1O_qGEYA1xPt.</url>
            <title>Stories by Andrey Matyash on Medium</title>
            <link>https://medium.com/@andrej-matyash?source=rss-dc35fb99fab2------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 23:32:38 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@andrej-matyash/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[ORFFA, ENGINEERING YOUR FEED SOLUTIONS IN AQUACULTURE]]></title>
            <link>https://andrej-matyash.medium.com/orffa-engineering-your-feed-solutions-in-aquaculture-e54d5c1766ad?source=rss-dc35fb99fab2------2</link>
            <guid isPermaLink="false">https://medium.com/p/e54d5c1766ad</guid>
            <dc:creator><![CDATA[Andrey Matyash]]></dc:creator>
            <pubDate>Mon, 17 May 2021 16:40:54 GMT</pubDate>
            <atom:updated>2021-05-17T16:40:54.390Z</atom:updated>
            <content:encoded><![CDATA[<p>WE CREATE, SELECT AND DELIVER FEED ADDITIVES</p><p>Orffa develops, customizes, sources and offers a complete range of feed additives and concepts for the animal nutrition and production markets. The escalating complexity of current challenges faced by the feed industry requires dedicated specialists, able to engineer speciﬁc applicable feed solutions. It is our mission to help the feed industry in ﬁnding such solutions. We invest in specialists, who are able to create, to select and to deliver feed additive solutions ﬁtting regional market needs throughout the world. Innovation, partnership, knowledge, quality, professionalism and global presence are key characteristics of Orffa’s dynamic growth strategy.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jTjlafCu6_OY1SVnliHq-A.jpeg" /></figure><p>FEED SOLUTIONS FAVORABLE FOR AQUATIC SPECIES<br>To enhance ﬁsh and shrimp health, performance, environment, welfare and proﬁtability</p><p>SELENIUM 4000 <br>Selenium 4000 is a new generation of organic selenium, with highly available and consistent selenium content due to its natural form.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8vaH5-W3tuObvakJMn8ptg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*S3pF5b696-M9aEWJ_4Z3IQ.jpeg" /></figure><p>ENERGY PLUS <br>As a nutritional emulsiﬁer, Energy Plus is designed to improve the fat digestibility, increase the energy uptake, to enhance animal performance, reduce feed cost and maximize proﬁtability.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*u42yDhevFw7OswjFn-uFxQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*m4jRJLWnU-wUChB4Z3kwxg.jpeg" /></figure><p>SMART <br>The trace minerals of SMART product range (copper, manganese and zinc) are hydroxy trace minerals with a crystalized structure and smart covalent bonds. The active minerals are gradually released in the gastro-intestinal tract, optimizing the absorption capacity.</p><p>TOXIN A / PLUS <br>For efﬁcient mycotoxin management, Toxin A / Plus are designed to effectively bind a broad spectrum of mycotoxins by combining different approaches.</p><p>ALLIIN PLUS <br>Alliin Plus is a combination of selected garlic and cinnamon sources containing concentrated levels of phytochemicals. Unique production proces allows for continued efﬁciency in order to boost the health status and counteract parasites.</p><p>BETA-KEY <br>Beta-Key is a non-hygroscopic betaine hydrochloride product. Methyl group donation, palatability and protection against osmotic stress beneﬁt ﬁsh performance.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ecld00UW3tMk-sXvogMarg.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fzzw0Lbeo-guH_LgQKcL2A.jpeg" /></figure><p>AMMOMIN <br>The clay mineral AmmoMIN absorbs nitrogen and heavy metals. This is applicable in water quality management and in diets as a feed additive.</p><p>BUTYCOAT / BUTY 90 <br>Butycoat and Buty 90 are butyrate concepts developed to provide energy for intestinal cells and stimulate a balanced gut health.</p><p>VITAMINS<br>Vitamin A <br>Vitamin AD3 <br>Vitamin B1 | B2 | B6 | B9 | B12 Nicotinamide | Niacin <br>D-Calpan<br>Vitamin C | coated | phosphate Vitamin D3<br>Vitamin E<br>Vitamin K3<br>Biotin<br>Choline Chloride<br>Inositol</p><p>CALSPORIN®<br>Stable Bacillus probiotic</p><p>BICAR®Z <br>Sodium bicarbonate</p><p>MACROGARD®<br>Beta-glucans</p><p>MINERALS AND TRACE ELEMENTS <br>Cobalt | Copper | Iron | Iodine | Manganese | Selenium | Zinc</p><p>AMINO ACIDS <br>Lysine | Threonine | Tryptophan | Valine | Methionine | Glycine | Cysteine | Carnitine | Taurine</p><p>PIGMENTS <br>Elovital Red — Canthaxantin 10% <br>Elovital Yellow — Apo Ester 10%</p><p>ACTIVEMOS<br>Mannan oligosaccharides</p><p>CAPSANTAL<br>Natural pigments</p><p>ROLE OF AQUACULTURE IN SUSTAINABILITY<br>Feed solutions are key for further development</p><p>Sustainability is a central pillar for aquaculture production. Aquaculture has a large production per unit area with a relatively small environmental footprint. Therefore, it is known to play a crucial role in meeting the challenge; how to feed a growing human population while reducing the environmental impact of food production. The aquaculture sector over the past 40 years had a continued increased market share, with current projections that consumption will increase to 109 million tons by 2030.</p><p>Ingredients and feeding strategies largely contribute to the sector’s sustainability. For instance, water quality is heavily inﬂuenced by feed ingredients which are not digested properly. Poor water quality conditions contain stressors that impair the ﬁsh health status and performance. Additionally, it is important to be aware of the impact of ingredients supplemented to compound feed to the global footprint. Together with you, Orffa delivers the solutions to contribute to a sustainable future with an extensive portfolio of feed additives.</p><p>RESEARCH AND INNOVATION<br>In the continuous development of Orffa’s products, research and innovation play a key role. We collaborate with a global network of universities, research institutes and consultants. We aim to provide the best feed solutions regarding optimal ﬁsh health, performance, welfare, environment and proﬁtability, in which a positive impact on the eco meet these ambitions.</p><p>ADD VALUE TO DIETS LOW IN FISHMEAL &amp; FISH OIL<br>Key products: Beta-Key, Butycoat, Energy Plus, Selenium 4000, Smart</p><p>For sustainability reasons, ﬁnite marine ingredients are increasingly replaced by alternatives, such as vegetable ingredients. Especially to carnivorous ﬁsh species, alternatives could negatively impact gut health.</p><p>ENVIRONMENTAL IMPACT<br>Key products: Alliin Plus, AmmoMIN, Energy Plus, Selenium 4000, Smart<br>Reduction of feces, improved digestibility, diminished feed spill and use of sustainable ingredients could all contribute to an improved environmental impact. We develop our products to maintain a high availability.</p><p>IMPACT ON (HUMAN) HEALTH<br>Key products: Alliin Plus, Butycoat, Selenium 4000, Smart, Toxin A / PLUS<br>Pathogens easily infect animals with a vulnerable immune system. A healthy ﬁsh population is essential to retain ﬁsh welfare and to increase proﬁtability. Phytogenics, mycotoxin binders, minerals, and butyrates all contribute in their own way to the health status of aquatic species. For example, the organic blend of garlic and cinnamon in Alliin Plus, is able to reduce the use of antibiotics. This is advantageous for human and ﬁsh health as it continues to reduce the antibiotic resistant pathogen population. The muscle or ﬁllet fortiﬁcation of feed additives, such as organic selenium, can both enhance the health status of ﬁsh as well as the human health.</p><p>FEED PROCESSING<br>Key products: AmmoMIN, Energy Plus, PelletAid Feed ingredients largely impact the ﬂowability of feed processing, which impacts the amount of energy used during production. Inclusion of feed additives to improve ﬂowability during a pelletizing process, could reduce production cost and decrease environmental impact.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e54d5c1766ad" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Bayesian Inferences allow us to find the best possible choice for a feed additive or a drug under…]]></title>
            <link>https://andrej-matyash.medium.com/bayesian-inferences-allow-us-to-find-the-best-possible-choice-for-a-feed-additive-or-a-drug-under-ad42e31a46e7?source=rss-dc35fb99fab2------2</link>
            <guid isPermaLink="false">https://medium.com/p/ad42e31a46e7</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[purchasing]]></category>
            <category><![CDATA[antibiotics]]></category>
            <category><![CDATA[feed-additives]]></category>
            <category><![CDATA[veterinary]]></category>
            <dc:creator><![CDATA[Andrey Matyash]]></dc:creator>
            <pubDate>Mon, 26 Aug 2019 10:05:31 GMT</pubDate>
            <atom:updated>2019-08-26T10:05:31.946Z</atom:updated>
            <content:encoded><![CDATA[<h3>Bayesian Inferences allow us to find the best possible choice for a feed additive or a drug under uncertainties.</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ltVOIbxeBG25PLq7j8jKLg.jpeg" /><figcaption>What is best possible choice under uncertainties ?</figcaption></figure><p>Bayesian Inferences allow us to find the best possible choice for a feed additive or a drug under uncertainties.</p><p>Holmes: “You <strong>see</strong>, but you do not <strong>observe</strong>. The distinction is clear. For example, you have frequently seen the steps which lead up from the hall to this room.”</p><p>Watson: “Frequently.”</p><p>Holmes: “How often?”</p><p>Watson: “Well, some hundreds of times.”</p><p>Holmes: “Then how many are there?”</p><p>Watson: “How many? I don’t know.”</p><p>Holmes: “Quite so! You have not observed. And yet you have seen. That is just my point. Now, I know that there are seventeen steps, because I have both seen and observed” (“A Scandal in Bohemia” in The Complete Sherlock Holmes. New York: Doubleday, 1927)</p><p>On a constant basis a nutritionist and veterinarian face challenges to find a best solution based on costs, production data or health issues in order to find a best solutions out of many options. With growing pressure from the public and government restriction in terms of use of antibiotics as growth promoters the risk assessment and choice of the right feed components and antibiotics is more than ever vital and must be based on clear evidence and statistical data and not on wishes and subjective opinions. How to decide on the purchase of a particular feed ingredient or antibiotic based data and evidence? The value of Bayesian probability model lies in its ability to aid the making of decisions under uncertainty.</p><p>The most sophisticated commonly used methods of decisions making based on risk assessment involve building statistical models from historical data. Yet such approaches are inadequate when risks are rare or novel, such as antibiotic resistant pathogens because there is no sufficient and relevant data on the subject.</p><p>Evidence based probabilistic models (called Bayesian) can provide a quantitative assessment of a best decision choice. Such models can capture the complex interdependencies between evidence of risk factors and can effectively combine data with expert judgement. The resulting models provide rigorous risk quantification as well as genuine decision for best choice of a feed ingredient or drug. Bayesian statistics integrate evidence in a systematic way, for decision-making purposes</p><p>Bayes’ Theorem is a method for interpreting probability in the view of a new evidences of the subject, this allow us constantly observe the situation. By definition a theorem is a mathematical statement has been proven true. Denying Bayes’ Theorem is like denying the theory of relativity.</p><p>Let’s consider a practical example from a veterinary medicine. Quite often Gentamicin is subscribed for pigs from 35 to 55 days old to treat bacterial infection causes by Streptococcus suis. Let’s consider is Gentamicin is the best choice? How can we prove it? This is where Bays’ Theorem can help us.</p><p>Let’s write symbol G as shorthand for Gentamicin. We assume that there is 50 % probability that Gentamicin is the best choice in a given situation. P (G) = 50 %. A Preevidence probability must be estimated in the Bayesian approach. In an ideal Bayesian world, we would wish to develop a preevidence probability that reflects complete ignorance about the issue at hand and then use Bayes’ theorem to update that probability in light of all the available evidence. But how is complete ignorance represented by probabilities? While some ignorance — that is, being in a situation where uncertainties are present — provides the very need for working with probabilities, what do we mean by complete ignorance? One solution is to say that complete ignorance is the situation in which you attach an equal probability to each possible outcome. For example, if there are two possible outcomes, only one of which can be true (but you don’t know which one), then complete ignorance about the outcome could be represented by attaching a 50 percent probability to each of them*(*Unwin, Dr Stephen D.. The Probability of God: A Simple Calculation That Proves the Ultimate Truth (Kindle Locations 655–661). The Crown Publishing Group. Kindle Edition.)</p><p>The negation of the hypothesis (G*) is, in a sense, the opposite hypothesis namely that Gentamicin is not the best choice. These two hypotheses are mutually exclusive and exhaustive set of hypotheses. Mutually exclusive means because G and G* cannot both be true; that is if one of these propositions is true, then other is false. They are called exhaustive because G and G* in combination, cover the entire range of possibilities — there is no third option. Yes, there are plenty of other antibiotics on a market but for purposes of this example let’s consider a simpler version of Bayesian Inferences with only two exclusive and exhaustive set of hypotheses.</p><p>To make quantitative risk assessment of a choice we need introduce a probability to each hypothesis.</p><p>P(G) — Prior probability — means the probability that Gentamicin is the best choice.</p><p>P(G|E) — Posterior probability — means the probability that Gentamicin is the best choice after the new evidence is accounted for.</p><p>So, in effect we have the before (prior) and after (posterior) probability: the probability assessed prior to the new evidence and the probability assessed after the new evidence is considered. In Bayesian jargon, these are known as prior and the posterior probability. P(G) — prior probability, and P(G|E) — posterior probability. Hypothesis G* that is NOT a best choice, also has its prior and posterior probabilities denoted P(G*) and P(G*|E), respectively.</p><p>P(E|G) represents the probability that the evidence E would have been produced if G were true. In other words, if it were the case that our choice of G is the best one, then P(E|G) is a priori probability that evidence E would have been produced. Similarly, with G*.</p><p>P(E|G*) = the probability that the evidence E would occur if G were false (i.e. if G* were true)</p><p>So here it is the Bays’ formula!</p><p>Pafter = Pbefore * P(E|G)/[Pbefore * P(E|G)] + [100%-Pbefore)*P(E|G*)</p><p>To simplify our calculations let’s introduce the evidence factor (EF). EV= P(E|G)/P(E|G*)</p><p>Pafter = Pbefore * EF /Pbefore * EF + 100%-Pbefore</p><p>EF (Evidence Factor) is defined by the equation EF= P(E|G)/P(E|G*). From the formula for EF, you can see that it is the factor by which a given piece of evidence is more likely to be produced if feed additive or antibiotic is the best choice. For example, if we state that EV=10 for a given piece of evidence, we are saying that that evidence is 10 times more likely to be produced if G were true the best choice. The greater its numerical value of EF, the more an item of evidence is assessed to point toward the right choice of a drug — and the smaller its value, the more the evidence points otherwise. The breakeven point is at D=1, indicating evidence that is Neutral choice; that is P(E|G) = P(E|G*). The advantage we have thus gained is that we need now only come up with a single number, that is D, for each piece of evidence rather that the two numbers P(E|G) and P(E|G*).</p><p>Here is screen shot from an excel sheet with bayesian inferences.</p><p>·</p><p>You can download this excel file <a href="https://www.dropbox.com/s/q6pek0p8k53axpm/data%20driven%20decisions%20based%20on%20Bayes%20Inference.xlsx?dl=0">here</a>.</p><p>Our estimation about use of Gentamicin based on the evidences to treat the Streptococcal infection went down from 50% to just 13 % ! Would you buy it in this situation ? Of course not !</p><p>The beauty of this bayesian approach is that we can apply it to any decision in our professional or personal life. It give us chance to observe any evidences and a solid probability number!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ad42e31a46e7" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>