Cross-modal representation alignment of molecular structure and perturbation-induced transcriptional profiles (2019-11-22T00:00:00.000000Z)

TL;DR

This work proposes a new cross-modal small molecule retrieval task, designed to force a model to learn to associate the structure of a small molecule with the transcriptional change it induces, developed formally as multi-view alignment problem, and presents a coordinated deep learning approach.

Abstract

Modeling the relationship between chemical structure and molecular activity is a key goal in drug development. Many benchmark tasks have been proposed for molecular property prediction, but these tasks are generally aimed at specific, isolated biomedical properties. In this work, we propose a new cross-modal small molecule retrieval task, designed to force a model to learn to associate the structure of a small molecule with the transcriptional change it induces. We develop this task formally as multi-view alignment problem, and present a coordinated deep learning approach that jointly optimizes representations of both chemical structure and perturbational gene expression profiles. We benchmark our results against oracle models and principled baselines, and find that cell line variability markedly influences performance in this domain. Our work establishes the feasibility of this new task, elucidates the limitations of current data and systems, and may serve to catalyze future research in small molecule representation learning.

Authors

Matthew B. A. McDermott

5 papers

S. G. Finlayson

1 papers

Alex V. Pickering

1 papers

TL;DR

Abstract

Authors

References42 items

Deep Learning Benchmarks on L1000 Gene Expression Data

Analyzing Learned Molecular Representations for Property Prediction

GNNExplainer: Generating Explanations for Graph Neural Networks

Dr.VAE: improving drug response prediction via modeling of drug perturbation effects

Unsupervised Multimodal Representation Learning across Medical Images and Reports

Systematic polypharmacology and drug repurposing via an integrated L1000-based Connectivity Map database mining

Deep Generative Modeling for Single-cell Transcriptomics

Harnessing the biological complexity of Big Data from LINCS gene expression signatures

Multi-modal Cycle-consistent Generalized Zero-Shot Learning

Machine learning in chemoinformatics and drug discovery.

Drug Repurposing Using Deep Embeddings of Gene Expression Profiles.

Relational inductive biases, deep learning, and graph networks

Semi-Supervised Biomedical Translation With Cycle Wasserstein Regression GANs

Triplet-Based Deep Hashing Network for Cross-Modal Retrieval

DeepProfile: Deep learning of cancer molecular profiles for precision medicine

Modeling polypharmacy side effects with graph convolutional networks

Cell-specific prediction and application of drug-induced gene expression profiles

Prediction Errors of Molecular Machine Learning Models Lower than Hybrid DFT Error.

A review of connectivity map and computational approaches in pharmacogenomics

Sampling Matters in Deep Embedding Learning

A Next Generation Connectivity Map: L1000 Platform and the First 1,000,000 Profiles

Information Retrieval and Text Mining Technologies for Chemistry.

Neural Message Passing for Quantum Chemistry

MoleculeNet: a benchmark for molecular machine learning

A Survey of Multi-View Representation Learning

Representing high throughput expression profiles via perturbation barcodes reveals compound targets

Molecular graph convolutions: moving beyond fingerprints

Deep Metric Learning Using Triplet Network

Making a Science of Model Search: Hyperparameter Optimization in Hundreds of Dimensions for Vision Architectures

Evaluation of Analytical Methods for Connectivity Map Data

Scikit-learn: Machine Learning in Python

Extended-Connectivity Fingerprints

The Connectivity Map: Using Gene-Expression Signatures to Connect Small Molecules, Genes, and Disease

Molecular dynamics simulations

SMILES. 2. Algorithm for generation of unique SMILES notation

Relations Between Two Sets of Variates

Massively multitask networks for drug

Siamese Neural Networks for One-Shot Image Recognition

a geometrical approach to identify differentially expressed genes, BMC bioinformatics 15, p

Hyperparameter optimization in hundreds of dimensions for vision architectures, Proceedings of the 30th International Conference on International Conference on Machine Learning - Volume 28 , p

Rdkit: Open-source cheminformatics

J . Leskovec , Gnnexplainer : Generating explanations for graph neural networks , Advances in neural information processing systems , 9244 ( 2019 )

Field of Study

Journal Information

Name

Page

Volume

Venue Information

Name

Type

URL

Alternate Names