Author

Kevin Bradley Dsouza

University of British ColumbiaORCID 0000-0002-2291-5143Unclaimed ORCID profile · follow to get notified of new preprints

Biology Disciplines

Click to filter author papers

Bioinformatics· 1 Genomics· 1

Signature Assays, Organisms & Methods

Click to filter · or open tag

Approach:deep learning· 2 Research Focus:dimensionality reduction· 2 Research Focus:gene regulatory network inference· 2 Approach:variational autoencoder· 2 Research Focus:benchmarking· 1 Assay:ChIP-seq· 1 Assay:DNA methylation· 1 Assay:Hi-C· 1

Frequent Co-Authors

Maxwell W. LibbrechtiD· 2 shared Vijay K. Bhargava· 2 shared Adam LiiD· 1 shared Alexandra Maslova· 1 shared Ediem Al-JiburyiD· 1 shared Matthias MerkenschlageriD· 1 shared

Activity by Year

2021-2022 · peak 1/yr

Co-Author Collaboration Network

6 top collaborators · 12 mutual co-author links

Click any collaborator node to inspect or filter shared papers

Focal author (center)Selected co-authorORCID co-authorUnlinked name

Curved chords = mutual co-authorship

2 shared papersGenomics

Maxwell W. LibbrechtORCID

Mutual co-authors in cluster

Author profile →

2 papers

Recent·Most cited

GenomicsNature CommunicationsPeer-reviewedJun 28, 2022

Hi-C-LSTM: Learning representations of chromatin contacts using a recurrent neural network identifies genomic drivers of conformation

Kevin Bradley Dsouza, Alexandra Maslova, Ediem Al-Jibury…·Maxwell W. Libbrecht

Hi-C-LSTM uses a recurrent neural network to learn low-dimensional representations of Hi-C data, enabling the identification of genomic elements that define chromatin conformation.

16 citations

Hi-Cdeep learning

BioinformaticsIEEE Transactions on Computational Biology and BioinformaticsPeer-reviewedMay 27, 2021

Latent representation of the human pan-celltype epigenome through a deep recurrent neural network

Kevin Bradley Dsouza, Adam Li, Vijay K. Bhargava, Maxwell W. Libbrecht

This paper introduces Epi-LSTM, a deep LSTM autoencoder that creates a pan-celltype epigenomic representation, outperforming existing methods in most cell types and providing smoother genomic representations.

5 citations

humanChIP-seq