• Home
  • Membership
    • Join the Pistoia Alliance
    • About Us
    • Vacancies
  • Projects
    • All Projects
    • AbVance
    • AI and ML
    • Chemical Safety Library
    • Controlled Substance Compliance Expert Community
    • DataFAIRy BioAssay
    • Digital Health Data in Clinical Studies
    • FAIR Implementation
    • HELM
    • Informed Consent Blockchain
    • Methods Database
    • Ontologies Mapping
    • SEED Project
    • UDM
    • User Experience for Life Sciences
    • New ideas
  • Events
    • October Virtual Conference Recordings
    • All events
    • Past Webinars
    • Submit an Event
  • News
  • Blog
  • Innovation
    • Submit an idea
    • President’s Startup Challenge
      • President’s Startup Challenge – for startups
      • Alumni 2019
      • Alumni 2018
      • Alumni 2017
      • Alumni 2016

  • Home
  • Membership
    • Join the Pistoia Alliance
    • About Us
    • Vacancies
  • Projects
    • All Projects
    • AbVance
    • AI and ML
    • Chemical Safety Library
    • Controlled Substance Compliance Expert Community
    • DataFAIRy BioAssay
    • Digital Health Data in Clinical Studies
    • FAIR Implementation
    • HELM
    • Informed Consent Blockchain
    • Methods Database
    • Ontologies Mapping
    • SEED Project
    • UDM
    • User Experience for Life Sciences
    • New ideas
  • Events
    • October Virtual Conference Recordings
    • All events
    • Past Webinars
    • Submit an Event
  • News
  • Blog
  • Innovation
    • Submit an idea
    • President’s Startup Challenge
      • President’s Startup Challenge – for startups
      • Alumni 2019
      • Alumni 2018
      • Alumni 2017
      • Alumni 2016
Contact Us

Search...

  • Menu

Pistoia Alliance October Virtual Conference: Describing Chemistry to Algorithms: Why Scientific Expertise Improves AI Accuracy

  • Tweet

This is part of the Pistoia Alliance Collaborate to Innovate Virtual Conference Week, October 19-23, 2020.  For more information about related events, please visit our online calendar.

 

Summary

More and better data is an obvious need, but have you considered how impactful descriptors can be on predictions? Join us to see how better descriptors can improve AI algorithm performance on predicting biological activity across multiple algorithmic approaches with over 150k compounds.

 

This case study by CAS and Dr. Alpha Lee (University of Cambridge) was recently peer reviewed and published in the Journal of Chemical Information and Modeling. It showcases the impact of better descriptors on prediction accuracy while also revealing unique insights beyond a standard descriptor that may get used today. If your AI initiatives aren’t meeting expectations, learn how better quality descriptors can immediately improve accuracy vs. existing fingerprints.

 

Learning Objectives

At the conclusion of this session, participants should be able to:
 

  • Learn how better quality descriptors can immediately improve accuracy vs. existing fingerprints.

 

Featured Topics

  • Better data is an obvious need for improved AI predictions, but have you considered the role of chemical descriptors?
  • Chemical descriptors often get overlooked, but this study demonstrates how drastically it can improve prediction rates, reveal hidden connections and insights.
  • See how better descriptors improve predictions across multiple algorithms.

 

Sponsored by: 

 

 

 

Yugal Sharma, PhD

Senior Director, Solutions, CAS

Dr. Sharma has over 15 years of experience in applying and managing data science and machine learning approaches to solving complex problems in the healthcare space. Yugal was a scientist for the National Institutes of Health (NIH), focusing on developing early disease detection algorithms. Since NIH, Yugal helped found a technology startup, followed by a business consulting startup. More recently, he applied his background as a government consultant focusing on analytics strategy for NIH and FDA clients. He has published several scientific articles, as well co-authoring a book chapter on the mining of Electronic Health Records to detect disease signals. He is currently the Sr. Director of Solutions for CAS Services which engineers customized approaches for your unique scientific information challenges – from AI/ ML initiatives, to faster workflows, and novel approaches to unlock value out of your scientific data. Yugal received his PhD in Biophysics from University of Cincinnati, where he graduated with honors.

 

Learn more: https://www.linkedin.com/in/yugalsharma/

 

 

Access the Recording

Posted on October 11, 2020 by Yasheaka Oakley
Categories: Pistoia Webinars

Tags: AI, Algorithms, CAS, chemistry, innovation, research, webinar


Events

20 Jan 2021

AI CoE Webinar: AI for Drug Repurposing

Book this event >
22 Jan 2021

Pistoia Alliance Discussion Group: Brexit

Book this event >
22 Jan 2021

Lab of the Future CoI Meeting: Lessons Learned in Creating Digital Labs

Book this event >
27 Jan 2021

Webinar: Informed Consent in Clinical Trials – Application of Blockchain Technology

Book this event >
Recent Posts
Bridging the Gap: Why Should UX’ers Care About Data?
01 Jan 2021

TweetData is Vital to Life Sciences   Data can...

Read More >
Learnings from judging the Code2Care Hackathon – the winners, ideas and themes and most importantly, focusing on the patient.
21 Oct 2020

TweetIn recent months I’ve had the privilege of seeing...

Read More >
Measuring UX Maturity with a UXLS Maturity Model
02 Oct 2020

TweetAs UX becomes a part of the DNA of...

Read More >
Design Thinking and UX are like two peas in a pod!
11 Jun 2020

TweetIn large companies, design sounds risky and introduces a...

Read More >
Follow Us
Active Projects
Informed Consent Blockchain
Digital health data in clinical studies
SEED Project
User Experience for Life Sciences
FAIR Implementation
DataFAIRy BioAssay
Methods Database
AI and ML
HELM
UDM
Sign up for Pistoia Alliance Communications

© Copyright Pistoia Alliance, Inc., 2020 | Contact us at info@pistoiaalliance.org | Terms of use | Privacy Statement

By clicking "Agree" you are agreeing to the following terms and conditions for the use of the UDM XML Schema definition.

This website intends to use cookies to improve the site and provide you with a better browsing experience. If you select "Continue" or continue to browse the site without customizing your choices, you agree to our use of cookies. Find out more in our Online Privacy Statement.

Continue More Info