Target Fellow; Postdoctoral Researcher - Machine Learning Drug Discovery

Target 2035 Fellow (Postdoctoral Researcher - Machine Learning for Drug Discovery)

Location:

On-site:
Toronto, Ontario, Canada (the Structural Genomics Consortium)

Online:
New York, NY (Columbia University)

Supervision:  Dr. Matthieu Schapira (SGC/UofT) and Dr. Mohammed AlQuraishi (Columbia University).

Position Overview
We are seeking a highly motivated Post-Doctoral Fellow to lead a cutting-edge project focused on advancing machine learning (ML) methodologies for drug discovery. This project addresses the fundamental challenges of data scarcity and generalization in drug discovery by leveraging large-scale datasets from DNA-encoded library (DEL) and affinity selection mass spectrometry (ASMS) platforms.

As a Mitacs Target 2035 post-doctoral fellow , you will work in an interdisciplinary team of computational chemists, ML experts and experimentalists at University of Toronto and Columbia University. The position will be physically located in the Schapira Lab at University of Toronto and will include regular meetings (weekly online and occasional on-site) with the team of Mohammed AlQuraishi at Columbia University.

Core Objectives
You will be responsible for executing two tightly integrated research objectives:

Defining the boundary between in and out of distribution for multi-modal binding affinity prediction deep learning models:  Fine-tune transformer-based affinity models on internal experimental ASMS/DEL high-throughput screening data and evaluate how prediction accuracy degrades as binding site distance from the training set increases — establishing the boundary for in-distribution predictions.

Identifying data gaps in training sets to guide experimental data generation:  Cluster the human proteome by binding site similarity and generate a ranked target list for experimental screening that maximizes coverage and fills data gaps in ML training sets. This list will guide proteome scale efforts at Target-2035 to generate experimental training sets that accelerate a breakthrough in AI-driven hit finding and optimization.

Role Responsibilities

Design and implement computational pipelines for proteome-scale binding site detection, featurization, and clustering using structural prediction models (Open Fold) and pocket mapping and embedding tools (Fpocket, Bio Emu, Uni Mol).

Fine-tune and benchmark transformer-based affinity prediction models (ex: AQAffinity, DrugCLIP) on ASMS and DEL screening datasets; evaluate generalization across structurally distant binding sites.

Develop and validate a distance metric for drug binding sites that quantifies the boundary between in-distribution and out-of-distribution prediction regimes.

Generate a dynamically updatable, prioritized target list for experimental screening that maximizes proteome coverage and fills data gaps in ML training sets.

Collaborate closely with the SGC experimental teams to align computational target selection with experimental tractability constraints.

Coordinate with the Schapira Lab (University of Toronto) and the AlQuraishi Lab (Columbia University / Open Fold) on AI-augmented computational chemistry and multi-modal model development integrating structural and binding data.

Present findings at internal Target 2035 consortium meetings and external scientific conferences; publish your work in peer-reviewed publications.

Qualifications

PhD in Computational Chemistry, Computer Science, Bioinformatics, or a related field.

Technical Expertise:
Strong background in representation learning, foundation models, and multi-fidelity optimization.

Domain Knowledge:
Familiarity with physics-based molecular modeling.

Skills:

Experience with large-scale ML training, handling "noisy" experimental data (like DEL)

Appointment, Training Program, and Research Environment
The fellowship is a full-time postdoctoral appointment for an initial one-year term, renewable for up to two additional years, subject to performance and program requirements.

This project operates within an open science framework, ensuring your work has a high impact on the global drug discovery community.

#J-18808-Ljbffr