Mobile Menu

Machine Learning for Disruptive Drug Development – a 3 part webinar series

The cost of drug R&D has increased by ten-fold in the last 30 years, but the success rate of candidate molecules passing clinical trials has almost halved in the same period. Technologies including advanced analytics and artificial intelligence have been much anticipated as the solution to the ongoing productivity conundrum.

And whilst recently we’ve seen a slight blip in the downward trajectory of R&D productivity, the first round of implementation and utilisation of machine learning in drug development has been far from miraculous.

In this virtual 3-part series our pharma leaders cast a critical eye on the success and failures of artificial intelligence and machine learning innovations over the few past years, to pinpoint where lessons can be learnt. We hear from them what the most pressing challenges are for those developing and employing ML approaches and how companies can evaluate and invest in the most appropriate tech to address their research questions.

Benefits of attending:

• Understand how a selection of top-tier pharma view the current state of AI and ML technologies for target identification and validation; small-molecule design and optimization; and predictive biomarker discovery

• Understand how the practical implementation of cutting-edge tools differs from their proof of-concept demonstration

• Hear real pharma examples of successes and failures in using ML to augment early-stage R&D

• Pinpoint where the remaining challenges and opportunities lie for the widespread adoption of AI/ML-driven approaches

Who should attend?

Drug developers working in target, pathway and molecule selection, developing predictive preclinical models or in translational research, either already employing predictive analytics and augmented intelligence or seeking to gain an insight into the practical implementation of such technologies.

Session one: Recent Innovations in Machine Learning for Target Identification and Validation

20th October 3pm BST/ 4pm CET/ 10am EST

• Grasp how computational approaches can leverage diverse omics datasets at scale to make in-silico target predictions

• Hear of recent advances in ML including natural language processing (NLP), for more powerful data mining of target association knowledge

• Understand the opportunities and challenges in using deep learning methods for the identification of potential therapeutic targets, including modelling genetic splicing variation


Shantanu Singh, Senior Group Leader in the Imaging Platform, Broad Institute

Tina Larson, President and COO, Recursion Pharma

Justin Boyd, Principal Scientist, Pfizer

Session two: Deep Learning for Small-Molecule Design and Optimization

27th October – 3pm GMT/ 4pm CET/ 10am EST

• Gain an insight into the advances in DL models used by a variety of pharma to predict absorption, distribution, metabolism, excretion, and toxicity (ADMETox) properties, in their lead optimization programs

Confirmed speakers:

Vishakha Sharma, Principal Data Scientist, Roche

Angelo Pugliese, Team Lead, Computational Chemistry – Artificial Intelligence, Cancer Research UK Beatson Institute – Drug Discovery Unit

Floriane Montanari, Research Scientist, Bayer

Govinda Bhisetti, Head of Computational Chemistry, Biogen

Session three: Machine Learning-based Predictive Biomarker Discovery in Preclinical Studies

3rd November – 3pm GMT/ 4pm CET/ 10am EST

• Hear of real examples where ML-based biomarker discovery and models to predict drug sensitivity based on pre-clinical data have demonstrated a significant improvement in clinical success rates.

• Learn where deep learning models are being applied to represent gene modules in biomarker discovery.

• Understand where progress needs to happen for clinical adoption of such biomarkers, including how to demonstrate the generalizability of the approach and apply objective approaches and measures to model training.

Confirmed speakers:

Shameer Khader, Ph.D, Senior Director (Data Science, Digital Health & Bioinformatics), AstraZeneca

George Lee, Digital Pathology Informatics Lead / Data Scientist, Bristol-Myers Squibb

Francisco Azuaje, Associate director, Data Science, UCB

Share this article