Hi [[ session.user.profile.firstName ]]

Applications of Biomedical Knowledge Graph for AI and Machine Learning

Biomedical knowledge graphs (BMKGs) link biomedical entities (such as diseases, proteins, and drugs) through certain defined relationships. They are important tools to computationally analyze the comprehensive body of biomedical knowledge.

In this webinar, Anton Yuryev, Biology Director at Elsevier will talk about the approaches and use of knowledge graph together with artificial Intelligence algorithms for various biomedical applications such as drug repurposing, personalized drug therapy, and personalized immunotherapy.

The talk will cover
- comparing different graph embedding techniques
- introducing a new graph embedding technique that uses patient OMICs data to calculate node activity using sub-network enrichment analysis
- how to use node activity to find likely disease mechanism in a patient and use this mechanism to predict personalized treatment or neoantigen vaccine design, or new target discovery

About speaker:
Dr. Anton Yuryev has PhD in Genetics from Johns Hopkins University where he discovered proteins physically linking gene transcription with mRNA processing in eukaryotic cells. He worked over 30 years in bioinformatics as Senior Scientist at InforMax, as Senior Bioinformatics Analyst at Orchid Cellmark, and as Senior Director of Application Science at Ariadne Genomics. Dr. Yuryev published over 50 scientific articles, edited four scientific books, authored algorithms for primer design and pathway analysis. He currently serves as Professional Services Director at Elsevier and responsible for development of targeted bioinformatics solutions using Elsevier proprietary software, knowledgebases and artificial intelligence in the areas of drug development, personalized precision medicine, agro- and synthetic biology.
Recorded Mar 11 2020 60 mins
Your place is confirmed,
we'll send you email reminders
Presented by
Anton Yuryev
Presentation preview: Applications of Biomedical Knowledge Graph for AI and Machine Learning

Network with like-minded attendees

  • [[ session.user.profile.displayName ]]
    Add a photo
    • [[ session.user.profile.displayName ]]
    • [[ session.user.profile.jobTitle ]]
    • [[ session.user.profile.companyName ]]
    • [[ userProfileTemplateHelper.getLocation(session.user.profile) ]]
  • [[ card.displayName ]]
    • [[ card.displayName ]]
    • [[ card.jobTitle ]]
    • [[ card.companyName ]]
    • [[ userProfileTemplateHelper.getLocation(card) ]]
  • Channel
  • Channel profile
  • DNA-encoded libraries in drug discovery Oct 14 2020 2:00 pm UTC 75 mins
    Dr. Alex Satz; Dr. Andreas Brunschweiger
    The development of DNA-encoded chemical libraries (DELs), an incredible chemistry advancement that makes it possible to contain trillions of compounds in a single test tube, is nothing short of revolutionary. DEL screening is now being used with increasing frequency by pharmaceutical companies for drug discovery.
    Dr. Alex Satz, senior director of DEL strategy and operations at WuXi AppTec, will provide an informative overview of DELs, followed by a presentation from Dr. Andreas Brunschweiger of TU Dortmund University on research demonstrating the role of cheminformatics and DNA-encoded chemistry in compound identification.
  • Reaxys逆合成予測ツールのご紹介 Oct 8 2020 5:00 am UTC 45 mins
    鈴木 直子
    低分子化合物の一連の合成反応のプランニング作成を助けるReaxysの新しい
    ソリューションとなる逆合成予測ツールをご紹介します。
    2018年にNature誌に発表されたM.H.S. SeglerとM. Waller(現Pending.AI)の
    画期的な研究に基づいて開発されたこの新しいソリューションは、最先端の
    ディープラーニング技術とReaxysの高品質な反応データを活用しています。
    このディープラーニング技術により、AIが化学反応を自己学習し、それに
    基いた逆合成予測によりアイデア創出等にご活用ください。
  • Using machine learning to extract chemical information from patents Oct 7 2020 2:00 pm UTC 60 mins
    Saber Akhondi, Principle NLP Scientist, Elsevier
    In commercial research and development projects, public disclosure of new chemical compounds and reactions often takes place in patents. Only a small proportion of these compounds are published in journals, usually a few years after the patent. Patent authorities make available the patents but do not provide systematic continuous chemical annotations. Different text-mining approaches exist to extract chemical information from patents but less attention has been given to relevancy of a compound in a patent. Relevancy of a compound to a patent is based on the patent’s context. A relevant compound plays a major role within a patent. Identification of relevant compounds reduces the size of the extracted data and improves the usefulness of patent resources (e.g. supports identifying the main compounds). Annotators of databases like Reaxys only annotate relevant compounds.
    Using the advanced technologies in Artificial intelligence (AI), Machine learning (ML) and Natural language processing (NLP), we have developed models to overcome these limitations. Through shared evaluation campaign we have also invited academic and industrial teams to further develop, improve and contribute to the domain of patent information extraction.

    The webinar will discuss:
    - The challenges of patent mining in the chemical domain
    - Chemical information extraction. From relevant document to relevant section to relevant information.
    - How to create a quality training set for machine learning in Chemistry
    - The ChEMU shared task for name entity and event extraction

    About speaker:

    Saber Akhondi obtained his MSc degree in Bioinformatics and Systems Biology from Chalmers University of Technology, Sweden. In 2011 he started as a PhD student within the biosemantics group in Erasmus Medical Center Rotterdam. He currently works at Elsevier as a Principle NLP Scientist where he applies NLP and machine learning techniques to extract information useful for large commercial and research communities.
  • Early Chemical Development on BMS-986095 for Hepatitis C Virus Recorded: Sep 23 2020 54 mins
    Changxia Yuan,Senior Research Investigator, Bristol Myers Squibb and Scott Newman, Customer Consultant, Elsevier
    Drug development is the process of bringing a new pharmaceutical drug to the market once a lead compound has been identified through the process of drug discovery. Chemical development of the synthetic/production process impact go/no go decisions of lead molecules. These include
    •Developing or inventing new routes for important process intermediates at scale
    •Optimizing reaction conditions for multiple routes based on novel reactants
    •Yield considerations at each step
    Each of these will be examined using a small molecule Hepatitis C nucleotide polymerase inhibitor that has been developed by BMS as a model.
  • COVID-19 search strategies in PharmaPendium: Focus on DMPK and DDIs Recorded: Sep 17 2020 47 mins
    Marnix Wieffer
    Many already approved drugs are currently being tested (repurposed) for prevention and treatment of COVID-19 infections and symptoms. To maximize repurposing and treatments success, you need to be fully informed on all regulatory DMPK, efficacy and drug safety data available on EMA and FDA approved drugs.

    Elsevier’s PharmaPendium provides access to full text searchable FDA/EMA drug approval Documents, manually extracted data, expert taxonomies and prediction tools. Thereby it is ideally positioned to support COVID-19 repurposing and treatments effort.

    In this webinar, PharmaPendium product manager Thomas Vargues and senior marketing manager drug safety Marnix Wieffer will show various PharmaPendium search strategies that are currently being used by pharmaceutical and healthcare customers.

    For this webinar they will focus on DMPK and DDI risk prediction workflows. For example:
    1.How does patient complications influence therapeutic PK profile and exposure?
    2. What DDIs can I expect now I am creating uncommon drug combinations?
    3.How to use Pharmacokinetic data to model exposure in COVID-19 patients?

    Getting answers to these critical questions can support drug repurposing efforts and reduce risk for patients.
  • COVID-19 search strategies in PharmaPendium: Focus on drug safety and efficacy Recorded: Sep 10 2020 47 mins
    Marnix Wieffer
    Many already approved drugs are currently being tested (repurposed) for prevention and treatment of COVID-19 infections and symptoms. To maximize repurposing and treatment success, you need to be fully informed on all regulatory DMPK, efficacy and drug safety data available on EMA and FDA approved drugs.

    Elsevier’s PharmaPendium provides access to full text searchable FDA/EMA drug approval Documents, manually extracted data, expert taxonomies and prediction tools. Thereby it is ideally positioned to support COVID-19 repurposing and treatments efforts.

    In this webinar PharmaPendium product manager Thomas Vargues and senior marketing manager drug safety Marnix Wieffer will show various PharmaPendium search strategies that are currently being used by pharmaceutical and healthcare customers. For this webinar they will focus on drug safety and efficacy workflows. For example:
    1.What adverse effects can I expect when dosing above approved therapeutic dose?
    2.What medication could aggravate COVID-19 disease symptoms and should be avoided?
    3.What approved drugs have been tested against COVID-19 related indications, endpoints, symptoms or targets?

    Getting answers to these critical questions can support drug repurposing efforts and reduce risk for patients.
  • Text mining as a solution:  Find disease-related genetic variation in literature Recorded: Sep 2 2020 44 mins
    George Jiang
    Genetic causes and associations of disease remain important areas of research so that our society can better prevent and treat diseases.
    Finding such genetic variation information within published literature remains a challenge due to the large volume and heterogeneity of articles, and the intense rate of publication in these fields.

    In this webinar we will cover:
    •How to find the context, not just the word
    •some use cases for text mining genetic variation data
    •show how we have developed a solution that enables researchers to quickly find genetic variations in the literature
    •Explore how you can leverage a semantic search solution to identify disease-related genetic variation information to address your research needs.

    Join George Jiang, Product Manager, Elsevier Text Mining solution and learn how to build your own solution with our text mining solution for searching for various genetic variation types.
  • Pharma data as an asset: how to become information centric Recorded: Jul 8 2020 59 mins
    Dr Martin Romacker
    Life Sciences and pharma are moving to a more data driven approach. Data is seen as an asset, yet not treated as such, with levels of investment not matching the value that reusable data will bring across the R&D value chain. Effective AI, ML and predictive outcomes are dependent on access to clean, reusable data and getting that clean data is often hard, time consuming and challenging.

    FAIR Data principles provide you with the perfect blueprint to overhaul your data processes.

    In this webinar, Dr Martin Romacker – Principal Scientist, Data and Information Architecture, Roche will walk us through how to access clean reusable data that will drive value across the R&D value chain and will outline areas where Roche is taking action to enable better and more effective data reuse.
  • Computer Programs for Semi-Automation of Evidence Synthesis Recorded: Jun 10 2020 60 mins
    Dr. Farhad Shokraneh, Research Fellow School of Medicine, University of Nottingham
    While all types of literature review are becoming reasonably more attractive for students, researchers, practitioners and policy makers the workload involved in all types of evidence synthesis should not be underestimated. Apart from standardization of procedures and methods, many organizations and collaborations started using computers to reduce the workload and save time in processing all types of reviews. Systematic reviews and meta-analyses, scoping reviews, rapid reviews, overviews, and realist reviews are only some members of review family that can benefit from using computer programs. The research, innovations, discussions, and skepticism around and surrounding the automation became so important that some of automation pioneers started International Collaboration for Automation of Systematic Reviews (ICASR) https://icasr.github.io/. The current webinar will also benefit the outcome of ICASR annual meetings.

    Despite emergence and listing of hundreds of tools in Systematic Review Toolbox (http://systematicreviewtools.com/), these software programs are underused. This webinar will introduce some of these programs alongside the evidence supporting their use and will provide a guide on how to choose the program, when to use them, what are their advantages and disadvantages, and why we should use them. There are automation tools for searching, screening, extracting data, analysis, and report writing. The presentation will also discuss the reasons for underusing problem and its solutions and will justify the fact that automation of evidence synthesis is still an idealist dream and why semi-automation of evidence synthesis is more realistic horizon for in the next decades.

    About the speaker: Farhad is an expert in systematic review (SR) methodology and automation, and manages the largest database (over 340 SRs) of schizophrenia trials. He uses Embase to provide search and consultancy to academicians, industry, clinicians and policy making teams around the world.
  • Reaxys Training Webinar 2-Structure Search_日本語版 Recorded: Jun 9 2020 33 mins
    Dr Marta da Pian
    Reaxysトレーニングビデオ:構造検索:Marvin JS/ ChemDrawエディタから反応検索まで
    ※日本語字幕付

    - MarvinJS/ChemDraw editor (drawing tools and checkbox panel)
    - Reaction search (product/reagent/catalyst, Atom mapping, Reacting center)
    - Diving into the Result panel
  • Chemistry Data for Systems Thinkers Recorded: May 26 2020 57 mins
    Paul Dockerty
    Systems thinking has become an essential part of modern medicinal chemistry and new drug development (1). Dealing with the increasing data volumes, information silos and low interoperability has become one of the biggest challenges to medicinal chemists when trying to take a holistic approach to identify the interactions and hidden connections within the organelles, cells, tissues, organisms. Often, we wonder:

    “Am I seeing the big picture without losing insight of the details?”

    In this webinar, we will discuss how to take a system approach to create new chemistry knowledge, to translate knowledge into useful applications and finally to be ready to face the unfolding world crises (2). The topics include
    - Systematic integration of biological and chemical data;
    - AI-ready data for synthesis route design and prediction;
    - Three practical examples using system thinking examples, including
    1. digitalization of chemistry knowledge in pharmaceutical industry,
    2. responding to COVID-19 pandemic using conscientious data excerption from literature,
    3. data readiness in green chemistry to support sustainability.

    Change management and education are inevitably critical to pursuing systems thinking approach, therefore we will talk about some best practices for pharmaceutical industry and educational system based on our learning's from the collaborative projects.

    (1) Systems Thinking for Medicinal Chemists, Jacobs Journal of Medicinal Chemistry, 2015, I (1),004
    (2) One-world chemistry and systems thinking, Nature Chemistry, 2016, 8, 393–398

    Paul Dockerty, PhD, is a Customer Engagement Manager in the Professional Services group at Elsevier, now responsible for supporting pharmaceutical customers in their digitalization journey. He is passionate about using data as a leverage to fight the natural resistance to change in digital transformations.
  • Addressing Questions & Unmet Needs in Melanoma Research and Treatment Recorded: May 19 2020 60 mins
    Marc Hurlbert; Tom Williams
    The landscape for melanoma research and treatment has rapidly changed over the last decade. Since 2011, the FDA has approved 12 new melanoma treatment regimens – including new classes of drugs that are molecularly targeted therapies (BRAF/MEK inhibitors), immune checkpoint inhibitors (anti CTLA-4, PD-1/PD-L1) and other immunotherapies (e.g. T-Vec). Scientists have also unraveled many of the genomic mutations found in the most common form, cutaneous melanoma, melanoma that arises primarily on sun-exposed areas of the skin. With these advances in research and treatment, the key unanswered questions have changed rapidly and existing preclinical models may not be sufficient to answer such questions surrounding immune checkpoint inhibition; resistance development, comparing to cuaneous melanoma, and how to improve early detection.

    Importantly, there are no models that accurately predict the patient journey. New models and additional research is needed to more fully represent all melanoma subtypes, stages, or treatment responses.

    About the speakers:
    Marc Hurlbert, Ph.D. Chief Science Officer, Melanoma Research Alliance. Marc is currently responsible for guiding MRA’s scientific strategy, overseeing the peer-reviewed grant-making program, and forging scientific collaborations. He has more than 18 years of nonprofit and grant-making experience focused on advancing medical research. Past work has included treatment and prevention strategies for breast cancer, lymphoma and multiple myeloma, as well as juvenile diabetes.

    Tom Williams, PhD, Life Sciences Professional Services Project Manager, Elsevier. Tom is a Life Sciences Knowledge Manager and Research Scientist. with extensive experience as an academic researcher in neurodegeneration and Alzheimer’s disease. He is also in skilled biophysical chemistry, dementia disorders, and biochemistry; and the author of many publications in the field of Alzheimer’s disease.
  • Optimizing clinical trial design with extracted efficacy data Recorded: May 7 2020 46 mins
    Marnix Wieffer
    Around 90% of the small molecule drugs that enter clinical trials do not make it to the market. Therefore, optimizing clinical trial design and reducing late stage failures are key priority for drug developers.

    With Phase II efficacy-related failure rates as high as 57%, many companies are seeking ways to improve their outcomes and reduce the climbing $2.6 billion costs to get one drug to market. As clinical trials become increasingly more complex and costly, is even more critical to mitigate the risk of failed clinical trials or arms due to suboptimal study design or poor efficacy

    Join us for this 45-minute webinar where Customer Consultant drug safety Jean-Dominique Pierret and Drug Safety Marketing Manager Dr. Marnix Wieffer will discuss how using PharmaPendium we can uncover critical information to make better more informed clinical development decisions.

    This will include in-depth information and demonstrations of how to leverage the comparative data in PharmaPendium to reduce the risk of late-stage failures. With a focus on efficacy, we will discuss how PharmaPendium enables you to:

    •Find efficacy weaknesses early
    •Identify the most appropriate preclinical models,
    •Improve success rates of Phase I and II clinical trial designs by optimising selection of sample size
    •Primary/secondary endpoint and study design and
    •Prepare for more effective regulatory reviews
  • Clinical and biochemical data-driven drug re-purposing for anti-infective drugs Recorded: Apr 30 2020 79 mins
    Andrey Khudoshin
    Drug repurposing has been shown to be advantageous for treating rare and common diseases. A data-driven drug repurposing approach may not only accelerate the time to reach the market but also helps in reducing costs and the clinical steps required, with the pre-existing knowledge of potential side-effects, special situations like age, gender or pregnancy, possibility to use combination with other drugs for more effective treatment, etc.

    In the webinar, five data-driven strategies for antiviral and antibiotic drug research and development will be discussed:
    • Search leveraging clinical data on drugs and biomolecules for treatment of related viral and bacterial disorders
    • Search for substances reported to be active against related viruses and microbes in patents and articles
    • Search for substances that interact with viral and bacterial proteins
    • Investigation of compounds affecting human proteins, involved in the viral life cycle
    • Assessment of the safety of drug candidate
    Above approaches will be demonstrated, utilizing the clinical data from Embase and experimental biochemical data from Reaxys Medicinal Chemistry

    About the speaker:
    Dr. Andrey Khudoshin holds a PhD in Chemistry from Lomonosov Moscow State University. Before starting a corporate career in the field of chemistry, biology and drug R&D in various international companies, he also completed his postdoctoral research on transformation of natural compounds to valuable substances, like raw materials for green chemistry and potential bioactive compounds. He joined Elsevier in 2015 and since then he is involved in the implementation of Elsevier's Life Science solutions, supporting increase the effectiveness of drug R&D.
  • Accelerate drug discovery by building and turning data into actionable insights Recorded: Apr 28 2020 57 mins
    Dr. Min Lu, Dr. Rosalind Sankey
    The prioritization of hits from large compound lists for further follow-up is a challenging task for medicinal chemists. During this step of drug discovery, multiple parameters such as synthetic accessibility, target specificity, physicochemical properties, and potential toxicities, in addition to desired biological activity, must be considered simultaneously. Increasing amounts of biological data are accumulating in the pharmaceutical industry and published literature (including journals and patents).

    However, data does not equal actionable information, and guidelines for appropriate data capture, harmonization, integration, mining, and visualization need to be established to fully harness its potential. Here, we describe ongoing efforts at Merck & Co. to structure data in the area of discovery chemistry. We are integrating complementary data from both internal and external data sources (Reaxys) into one, and will demonstrate how this well-curated database facilitates compound set design, tool compound selection, target deconvolution in phenotypic screening, and predictive model building (e.g. target prediction).

    Early in the discovery process, chemists select a subset of compounds for further research, often from many viable candidates. These decisions determine the success of a discovery campaign, and ultimately what kind of drugs are developed and marketed to the public. We present our findings in the context of complex problem solving and decision theory, and discuss the implications on drug discovery.

    Speakers
    Min Lu, Ph.D. Merck Research Laboratories, Boston, USA
    Rosalind Sankey, Ph.D. Elsevier Chemistry Solutions, Frankfurt, Germany
  • Finding novel lead compounds in pesticide discovery inspired by pharma research Recorded: Apr 8 2020 49 mins
    Dr Maria Shkrob, Dr Frederik van den Broek
    The use of high throughput (HTP) methodologies for supporting discovery and development of new agrochemical products opens up new opportunities to test many new compounds potentially acting on biological targets in various organisms. Finding new lead compounds which might act as a new pesticide can sometimes be a lengthy process; we present a method which can provide lead compounds by using the breath of information available from pharmaceutical research.

    This webinar will give an overview of the chemical and biological informatics methods and data used to map compounds active against biological targets in parasites in humans to fungal targets. The webinar will then also explore how to arrive at insights in structure activity relationships and freedom to operate in the chemical space for these candidate compounds. Thereby demonstrating how findings from pharmaceutical research can be transferred to fungal research.
  • Applications of Biomedical Knowledge Graph for AI and Machine Learning Recorded: Mar 11 2020 60 mins
    Anton Yuryev
    Biomedical knowledge graphs (BMKGs) link biomedical entities (such as diseases, proteins, and drugs) through certain defined relationships. They are important tools to computationally analyze the comprehensive body of biomedical knowledge.

    In this webinar, Anton Yuryev, Biology Director at Elsevier will talk about the approaches and use of knowledge graph together with artificial Intelligence algorithms for various biomedical applications such as drug repurposing, personalized drug therapy, and personalized immunotherapy.

    The talk will cover
    - comparing different graph embedding techniques
    - introducing a new graph embedding technique that uses patient OMICs data to calculate node activity using sub-network enrichment analysis
    - how to use node activity to find likely disease mechanism in a patient and use this mechanism to predict personalized treatment or neoantigen vaccine design, or new target discovery

    About speaker:
    Dr. Anton Yuryev has PhD in Genetics from Johns Hopkins University where he discovered proteins physically linking gene transcription with mRNA processing in eukaryotic cells. He worked over 30 years in bioinformatics as Senior Scientist at InforMax, as Senior Bioinformatics Analyst at Orchid Cellmark, and as Senior Director of Application Science at Ariadne Genomics. Dr. Yuryev published over 50 scientific articles, edited four scientific books, authored algorithms for primer design and pathway analysis. He currently serves as Professional Services Director at Elsevier and responsible for development of targeted bioinformatics solutions using Elsevier proprietary software, knowledgebases and artificial intelligence in the areas of drug development, personalized precision medicine, agro- and synthetic biology.
  • Drug repurposing for rare diseases: an integrated data driven approach Recorded: Feb 27 2020 53 mins
    Jabe Wilson; Megan Golden
    Pharma R&D for rare diseases is, itself, rare, because patient populations are perceived as insufficient to deliver ROI. Drug repurposing eliminates the expensive process of discovering a completely new compound, shortens the time that is needed to reach the patients, and ensures a higher success rate.

    Elsevier and Pistoia Alliance organized a drug-repurposing datathon, with Cures Within Reach and Mission: Cure being the consulting organizations. The objective was to identify repurposable drug candidates for chronic pancreatitis – a rare disease that affects about 1 million people globally, and currently doesn’t have an approved treatment. As a result, this datathon identified 4 drug candidates in 30-60 days. They were reviewed and approved by the expert panel, pending further clinical trials by Mission:Cure.

    This webinar will talk about this unique non-profit and private collaborative datathon, using Entellect, an AI-powered technology platform for identifying repurposable drug candidates for chronic pancreatitis.

    The topics will include:
    - Introduction to datathon
    - Predictive analytics for drug repurposing - needs and challenges
    - Established strategy and workflow
    - Outcome and impact
    - Update to the clinical trial progress

    About speaker:
    Jabe Wilson, Global Commercial Director, Data and Analytics, Elsevier
    Megan Golden, co-founder and co-director, Mission: Cure
  • Predicting drug-drug interactions to reduce adverse event risk Recorded: Feb 26 2020 49 mins
    Marnix Wieffer
    Adverse drug reactions (ADRs) are a serious problem worldwide. One reason for the increase in ADRs is the growth in prescription use—especially among aging populations where drug–drug interactions (DDIs) are more likely. Currently, 9 percent of Americans over age 55 take 10 or more prescription drugs, which greatly increases the likelihood of DDIs and ADRs.

    Identifying potential drug-drug interaction risk is a key priority for pharmaceutical manufacturers and regulatory authorities. To minimize health risks to clinical trial subjects and patients, assessments should be performed as early as possible in development.
    Join solution marketing manager Dr Marnix Wieffer for this webinar where he will discuss outstanding issues with predicting Drug-drug interaction risk and possible solutions. We run through a couple of live examples that show how PharmaPendium is supporting Drug-drug interaction risk prediction.
    Using PharmaPendium we will investigate
    •What enzymes and transporters act on my drug of interest?
    •What is the DDI risk for drugs that are substrates CYP2D6?
    •What is the DDI risk of my drug under development with Antiarrhythmic drugs?
  • Using machine learning to identify adverse events from scientific literature Recorded: Feb 19 2020 61 mins
    Umesh Nandal
    Information found in the biomedical literature is a significant source for tracking and reporting adverse drug reactions (ADR). The EMA and FDA have both mandated that market authorization holders maintain active screening of literature for any mentions of ADRs related to their drugs or other medicinal products. Given the increasing amount of literature, manual screening, reviewing and monitoring literature costs more time, money and creates an additional compliance risk. Using the advanced technologies in Artificial intelligence (AI), Machine learning (ML) and Natural language processing (NLP), we have developed models to identify Adverse events (AE) in the literature, which can save considerable time and effort in large-scale analysis and in integrating data from multiple diverse information sources.

    This webinar will discuss:
    - the challenges of literature mining using AI
    - the Biomedical Named Entity Recognition (BNER) and its advantages for information extraction tasks
    - how to create a quality training set for machine learning
    - the experiment outcome and further applications

    About speaker:

    Umesh Nandal, PhD, is the Principal Machine Learning & NLP scientist in Content Transformation (CT) department at Elsevier. With a background in Chemistry and computational biology, Umesh is applying state-of-the-art methods in ML and NLP to improve or build new life science products of Elsevier that can help researchers in getting correct answers to their questions quickly. Prior to joining Elsevier, he used various ML and computational approaches to analyse molecular data generated from high-throughput technologies to understand biological processes in healthy and diseased organisms. During his PhD, he intensively worked on the comparison of mouse models with humans by building a network based integration method that can compare their biological networks.
Webinars supporting your innovation, compliance needs, and safety
Realize the potential of data-driven drug discovery and development with solutions that help you increase research efficiency and ensure a thriving R&D pipeline.

Embed in website or blog

Successfully added emails: 0
Remove all
  • Title: Applications of Biomedical Knowledge Graph for AI and Machine Learning
  • Live at: Mar 11 2020 3:00 pm
  • Presented by: Anton Yuryev
  • From:
Your email has been sent.
or close