New Frontiers of AI for Drug Discovery and Development

NeurIPS 2023 Workshop @ New Orleans, Louisiana, USA

Room 242 @ Ernest N. Morial Convention Center

December 15, 2023

Workshop Schedule

All times listed are in Central Standard Time (CST, UTC-6, New Orleans Local Time).

Time	Event
8:15 - 8:25	Opening Remarks
8:25 - 9:05	Roundtable Discussion
9:05 - 9:40	Invited Talk: Wengong Jin DSMBind: SE(3) denoising score matching for unsupervised binding energy prediction and nanobody design
9:40 - 10:00	Oral Presentations *Jesus de la Fuente: Towards a more inductive world for drug repurposing approaches Julian Alverio*: In vitro validated antibody design against multiple therapeutic antigens using generative inverse folding
10:00 - 10:30	Coffee Break
10:30 - 11:05	Invited Talk: Marinka Zitnik Foundation Models for Molecular Drug Design and Clinical Drug Development
11:05 - 11:25	Oral Presentations *Kieran Didi: A framework for conditional diffusion modelling with applications in motif scaffolding for protein design Xuefeng Liu*: DrugImprover: Utilizing Reinforcement Learning for Multi-Objective Alignment in Drug Optimization
11:25 - 12:00	Invited Talk: Haoda Fu LLM Is Not All You Need. Generative AI on Smooth Manifolds
12:00 - 1:20	Lunch Break
1:20 - 2:25	Poster Session
2:25 - 3:00	Invited Talk: Michael Bronstein Harnessing geometry for molecular design
3:00 - 3:30	Coffee Break
3:30 - 4:05	Invited Talk: Jian Tang Diffusion Models for Molecular Structure Prediction
4:05 - 4:40	Invited Talk: Iya Khalil Decoding Biology with AI and High-Throughput Biology
4:40 - 5:10	Award Ceremony *Denis Tarasov: Offline RL for generative design of protein binders Lucian Chan*: Embracing assay heterogeneity with neural processes for markedly improved bioactivity predictions
5:10 - 5:20	Concluding Remarks

Oral Presentation Schedule

All times listed are in Central Standard Time (CST, UTC-6, New Orleans Local Time).

Time	Paper Title
9:40	Towards a more inductive world for drug repurposing approaches
9:50	In vitro validated antibody design against multiple therapeutic antigens using generative inverse folding
11:05	A framework for conditional diffusion modelling with applications in motif scaffolding for protein design
11:15	DrugImprover: Utilizing Reinforcement Learning for Multi-Objective Alignment in Drug Optimization
4:40	Offline RL for generative design of protein binders
4:55	Embracing assay heterogeneity with neural processes for markedly improved bioactivity predictions

Accepted Papers

Model-free selective inference and its applications to drug discovery (Poster)

RetroBridge: Modeling Retrosynthesis with Markov Bridges (Poster)

Sample-efficient Antibody Design through Protein Language Model for Risk-aware Batch Bayesian Optimization (Poster)

Embracing assay heterogeneity with neural processes for markedly improved bioactivity predictions (Best Paper Award; Oral)

Offline RL for generative design of protein binders (Best Student Paper Award; Oral)

The neural scaling laws for phenotypic drug discovery (Poster)

TacoGFN: Target Conditioned GFlowNet for Drug Design (Poster)

Towards a more inductive world for drug repurposing approaches (Oral)

Automating reward function configuration for drug design (Poster)

De novo design of antibody heavy chains with SE(3) diffusion (Poster)

Leap: molecular synthesisability scoring with intermediates (Poster)

Machine Learning Guided AQFEP: A Fast & Efficient Absolute Free Energy Perturbation Solution for Virtual Screening (Poster)

Large-scale Pretraining Improves Sample Efficiency of Active Learning based Molecule Virtual Screening (Poster)

Data-Efficient Molecular Generation with Hierarchical Textual Inversion (Poster)

Learning Scalar Fields for Molecular Docking with Fast Fourier Transforms (Poster)

Gotta be SAFE: A New Framework for Molecular Design (Poster)

TopoPool: An Adaptive Graph Pooling Layer for Extracting Molecular and Protein Substructures (Poster)

CongFu: Conditional Graph Fusion for Drug Synergy Prediction (Poster)

AlphaFold Meets Flow Matching for Generating Protein Ensembles (Poster)

Do chemical language models provide a better compound representation? (Poster)

CryoSTAR: Cryo-EM Heterogeneous Reconstruction of Atomic Models with Structural Regularization (Poster)

PharmacoNet: Accelerating Large-Scale Virtual Screening by Deep Pharmacophore Modeling (Poster)

FragXsiteDTI: an interpretable transformer-based model for drug-target interaction prediction (Poster)

Compositional Deep Probabilistic Models of DNA Encoded Libraries (Poster)

Graph Neural Bayesian Optimization for Virtual Screening (Poster)

PiNUI: A Dataset of Protein-Protein Interactions for Machine Learning (Poster)

PDB-Struct: A Comprehensive Benchmark for Structure-based Protein Design (Poster)

MoleculeGPT: Instruction Following Large Language Models for Molecular Property Prediction (Poster)

Protein Language Model-Powered 3D Ligand Binding Site Prediction from Protein Sequence (Poster)

Role of Structural and Conformational Diversity for Machine Learning Potentials (Poster)

DiffDock-Pocket: Diffusion for Pocket-Level Docking with Sidechain Flexibility (Poster)

Explaining Drug Repositioning: A Case-Based Reasoning Graph Neural Network Approach (Poster)

Multitask-Guided Self-Supervised Tabular Learning for Patient-Specific Survival Prediction (Poster)

TrustAffinity: accurate, reliable and scalable out-of-distribution protein-ligand binding affinity prediction using trustworthy deep learning (Poster)

Generalist Equivariant Transformer Towards 3D Molecular Interaction Learning (Poster)

Scalable Normalizing Flows Enable Boltzmann Generators for Macromolecules (Poster)

Removing Biases from Molecular Representations via Information Maximization (Poster)

Protein Language Models Enable Accurate Cryptic Ligand Binding Pocket Prediction (Poster)

VN-EGNN: Equivariant Graph Neural Networks with Virtual Nodes Enhance Protein Binding Site Identification (Poster)

Neurosymbolic AI Reveals Biases and Limitations in ML-Driven Drug Discovery (Poster)

DGFN: Double Generative Flow Networks (Poster)

Tree Search-Based Evolutionary Bandits for Protein Sequence Optimization (Poster)

PoseCheck: Generative Models for 3D Structure-based Drug Design Produce Unrealistic Poses (Poster)

Leveraging expert feedback to align proxy and ground truth rewards in goal-oriented molecular generation (Poster)

AbLEF: Antibody Language Ensemble Fusion for thermodynamically empowered property predictions (Poster)

Ab-DeepGA: A generative modeling framework leveraging deep learning for antibody affinity tuning (Poster)

Hit Expansion Driven By Machine Learning (Poster)

MolSiam: Simple Siamese Self-supervised Representation Learning for Small Molecules (Poster)

On Modelability and Generalizability: Are Machine Learning Models for Drug Synergy Exploiting Artefacts and Biases in Available Data? (Poster)

In vitro validated antibody design against multiple therapeutic antigens using generative inverse folding (Oral)

Evaluating Zero-Shot Scoring for In Vitro Antibody Binding Prediction with Experimental Validation (Poster)

GraphPrint: Extracting Features from 3D Protein Structure for Drug Target Affinity Prediction (Poster)

Online Learning of Optimal Prescriptions under Bandit Feedback with Unknown Contexts (Poster)

All You Need is LOVE: Large Optimized Vector Embeddings Network for Drug Repurposing (Poster)

ExPT: Synthetic Pretraining for Few-Shot Experimental Design (Poster)

SALSA: Semantically-Aware Latent Space Autoencoder (Poster)

A framework for conditional diffusion modelling with applications in motif scaffolding for protein design (Oral)

DrugImprover: Utilizing Reinforcement Learning for Multi-Objective Alignment in Drug Optimization (Oral)

Inpainting Protein Sequence and Structure with ProtFill (Poster)

Identifying regularization schemes that make feature attributions faithful (Poster)