Cargando…

Multinomial Convolutions for Joint Modeling of Regulatory Motifs and Sequence Activity Readouts

A common goal in the convolutional neural network (CNN) modeling of genomic data is to discover specific sequence motifs. Post hoc analysis methods aid in this task but are dependent on parameters whose optimal values are unclear and applying the discovered motifs to new genomic data is not straight...

Descripción completa

Detalles Bibliográficos
Autores principales: Park, Minjun, Singh, Salvi, Khan, Samin Rahman, Abrar, Mohammed Abid, Grisanti, Francisco, Rahman, M. Sohel, Samee, Md. Abul Hassan
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9498894/
https://www.ncbi.nlm.nih.gov/pubmed/36140783
http://dx.doi.org/10.3390/genes13091614
_version_ 1784794873745899520
author Park, Minjun
Singh, Salvi
Khan, Samin Rahman
Abrar, Mohammed Abid
Grisanti, Francisco
Rahman, M. Sohel
Samee, Md. Abul Hassan
author_facet Park, Minjun
Singh, Salvi
Khan, Samin Rahman
Abrar, Mohammed Abid
Grisanti, Francisco
Rahman, M. Sohel
Samee, Md. Abul Hassan
author_sort Park, Minjun
collection PubMed
description A common goal in the convolutional neural network (CNN) modeling of genomic data is to discover specific sequence motifs. Post hoc analysis methods aid in this task but are dependent on parameters whose optimal values are unclear and applying the discovered motifs to new genomic data is not straightforward. As an alternative, we propose to learn convolutions as multinomial distributions, thus streamlining interpretable motif discovery with CNN model fitting. We developed MuSeAM (Multinomial CNNs for Sequence Activity Modeling) by implementing multinomial convolutions in a CNN model. Through benchmarking, we demonstrate the efficacy of MuSeAM in accurately modeling genomic data while fitting multinomial convolutions that recapitulate known transcription factor motifs.
format Online
Article
Text
id pubmed-9498894
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-94988942022-09-23 Multinomial Convolutions for Joint Modeling of Regulatory Motifs and Sequence Activity Readouts Park, Minjun Singh, Salvi Khan, Samin Rahman Abrar, Mohammed Abid Grisanti, Francisco Rahman, M. Sohel Samee, Md. Abul Hassan Genes (Basel) Article A common goal in the convolutional neural network (CNN) modeling of genomic data is to discover specific sequence motifs. Post hoc analysis methods aid in this task but are dependent on parameters whose optimal values are unclear and applying the discovered motifs to new genomic data is not straightforward. As an alternative, we propose to learn convolutions as multinomial distributions, thus streamlining interpretable motif discovery with CNN model fitting. We developed MuSeAM (Multinomial CNNs for Sequence Activity Modeling) by implementing multinomial convolutions in a CNN model. Through benchmarking, we demonstrate the efficacy of MuSeAM in accurately modeling genomic data while fitting multinomial convolutions that recapitulate known transcription factor motifs. MDPI 2022-09-08 /pmc/articles/PMC9498894/ /pubmed/36140783 http://dx.doi.org/10.3390/genes13091614 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Park, Minjun
Singh, Salvi
Khan, Samin Rahman
Abrar, Mohammed Abid
Grisanti, Francisco
Rahman, M. Sohel
Samee, Md. Abul Hassan
Multinomial Convolutions for Joint Modeling of Regulatory Motifs and Sequence Activity Readouts
title Multinomial Convolutions for Joint Modeling of Regulatory Motifs and Sequence Activity Readouts
title_full Multinomial Convolutions for Joint Modeling of Regulatory Motifs and Sequence Activity Readouts
title_fullStr Multinomial Convolutions for Joint Modeling of Regulatory Motifs and Sequence Activity Readouts
title_full_unstemmed Multinomial Convolutions for Joint Modeling of Regulatory Motifs and Sequence Activity Readouts
title_short Multinomial Convolutions for Joint Modeling of Regulatory Motifs and Sequence Activity Readouts
title_sort multinomial convolutions for joint modeling of regulatory motifs and sequence activity readouts
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9498894/
https://www.ncbi.nlm.nih.gov/pubmed/36140783
http://dx.doi.org/10.3390/genes13091614
work_keys_str_mv AT parkminjun multinomialconvolutionsforjointmodelingofregulatorymotifsandsequenceactivityreadouts
AT singhsalvi multinomialconvolutionsforjointmodelingofregulatorymotifsandsequenceactivityreadouts
AT khansaminrahman multinomialconvolutionsforjointmodelingofregulatorymotifsandsequenceactivityreadouts
AT abrarmohammedabid multinomialconvolutionsforjointmodelingofregulatorymotifsandsequenceactivityreadouts
AT grisantifrancisco multinomialconvolutionsforjointmodelingofregulatorymotifsandsequenceactivityreadouts
AT rahmanmsohel multinomialconvolutionsforjointmodelingofregulatorymotifsandsequenceactivityreadouts
AT sameemdabulhassan multinomialconvolutionsforjointmodelingofregulatorymotifsandsequenceactivityreadouts