Cargando…

Cross-modal representation alignment of molecular structure and perturbation-induced transcriptional profiles

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 cr...

Descripción completa

Detalles Bibliográficos
Autores principales: Finlayson, Samuel G., McDermott, Matthew B.A., Pickering, Alex V., Lipnick, Scott L., Kohane, Isaac S.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8397230/
https://www.ncbi.nlm.nih.gov/pubmed/33691024
_version_ 1783744568864276480
author Finlayson, Samuel G.
McDermott, Matthew B.A.
Pickering, Alex V.
Lipnick, Scott L.
Kohane, Isaac S.
author_facet Finlayson, Samuel G.
McDermott, Matthew B.A.
Pickering, Alex V.
Lipnick, Scott L.
Kohane, Isaac S.
author_sort Finlayson, Samuel G.
collection PubMed
description 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.
format Online
Article
Text
id pubmed-8397230
institution National Center for Biotechnology Information
language English
publishDate 2021
record_format MEDLINE/PubMed
spelling pubmed-83972302021-08-27 Cross-modal representation alignment of molecular structure and perturbation-induced transcriptional profiles Finlayson, Samuel G. McDermott, Matthew B.A. Pickering, Alex V. Lipnick, Scott L. Kohane, Isaac S. Pac Symp Biocomput Article 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. 2021 /pmc/articles/PMC8397230/ /pubmed/33691024 Text en https://creativecommons.org/licenses/by-nc/4.0/Open Access chapter published by World Scientific Publishing Company and distributed under the terms of the Creative Commons Attribution Non-Commercial (CC BY-NC) 4.0 License.
spellingShingle Article
Finlayson, Samuel G.
McDermott, Matthew B.A.
Pickering, Alex V.
Lipnick, Scott L.
Kohane, Isaac S.
Cross-modal representation alignment of molecular structure and perturbation-induced transcriptional profiles
title Cross-modal representation alignment of molecular structure and perturbation-induced transcriptional profiles
title_full Cross-modal representation alignment of molecular structure and perturbation-induced transcriptional profiles
title_fullStr Cross-modal representation alignment of molecular structure and perturbation-induced transcriptional profiles
title_full_unstemmed Cross-modal representation alignment of molecular structure and perturbation-induced transcriptional profiles
title_short Cross-modal representation alignment of molecular structure and perturbation-induced transcriptional profiles
title_sort cross-modal representation alignment of molecular structure and perturbation-induced transcriptional profiles
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8397230/
https://www.ncbi.nlm.nih.gov/pubmed/33691024
work_keys_str_mv AT finlaysonsamuelg crossmodalrepresentationalignmentofmolecularstructureandperturbationinducedtranscriptionalprofiles
AT mcdermottmatthewba crossmodalrepresentationalignmentofmolecularstructureandperturbationinducedtranscriptionalprofiles
AT pickeringalexv crossmodalrepresentationalignmentofmolecularstructureandperturbationinducedtranscriptionalprofiles
AT lipnickscottl crossmodalrepresentationalignmentofmolecularstructureandperturbationinducedtranscriptionalprofiles
AT kohaneisaacs crossmodalrepresentationalignmentofmolecularstructureandperturbationinducedtranscriptionalprofiles