ChromaFold predicts the 3D contact map from single-cell chromatin accessibility.

Autor:	Gao VR; Computational and Systems Biology Program, Memorial Sloan Kettering Cancer Center, New York, NY, USA.; Tri-Institutional Program in Computational Biology and Medicine, New York, NY, USA., Yang R; Computational and Systems Biology Program, Memorial Sloan Kettering Cancer Center, New York, NY, USA.; Tri-Institutional Program in Computational Biology and Medicine, New York, NY, USA., Das A; University of Washington, Seattle, WA, USA., Luo R; Developmental Biology Program, Sloan Kettering Institute, New York, NY, USA., Luo H; Molecular Pharmacology Program, Experimental Therapeutics Center and Center for Stem Cell Biology, Memorial Sloan Kettering Cancer Center, New York, NY, USA., McNally DR; Caryl and Israel Englander Institute for Precision Medicine, Institute for Computational Biomedicine, Weill Cornell Medicine, Cornell University, New York, NY, USA., Karagiannidis I; Division of Hematology and Medical Oncology, Department of Medicine, Weill Cornell Medical College, New York, NY, USA., Rivas MA; Division of Hematology and Medical Oncology, Department of Medicine, Weill Cornell Medical College, New York, NY, USA., Wang ZM; Howard Hughes Medical Institute and Immunology Program, Sloan Kettering Institute and Ludwig Center at Memorial Sloan Kettering Cancer Center, New York, NY, USA., Barisic D; Division of Hematology and Medical Oncology, Department of Medicine, Weill Cornell Medical College, New York, NY, USA., Karbalayghareh A; Computational and Systems Biology Program, Memorial Sloan Kettering Cancer Center, New York, NY, USA., Wong W; Computational and Systems Biology Program, Memorial Sloan Kettering Cancer Center, New York, NY, USA.; Tri-Institutional Program in Computational Biology and Medicine, New York, NY, USA., Zhan YA; Center for Epigenetics Research, Memorial Sloan Kettering Cancer Center, New York, NY, USA., Chin CR; Division of Hematology and Medical Oncology, Department of Medicine, Weill Cornell Medical College, New York, NY, USA., Noble W; University of Washington, Seattle, WA, USA., Bilmes JA; University of Washington, Seattle, WA, USA., Apostolou E; Sanford I Weill department of Medicine, Sandra and Edward Meyer Cancer center, Weill Cornell Medicine, New York, NY, USA., Kharas MG; Molecular Pharmacology Program, Experimental Therapeutics Center and Center for Stem Cell Biology, Memorial Sloan Kettering Cancer Center, New York, NY, USA., Béguelin W; Division of Hematology and Medical Oncology, Department of Medicine, Weill Cornell Medical College, New York, NY, USA., Viny AD; Departments of Medicine, Division of Hematology & Oncology, and of Genetics & Development, Columbia Stem Cell Initiative, Herbert Irving Comprehensive Cancer Center, Columbia University Irving Medical Center, New York, NY, USA., Huangfu D; Developmental Biology Program, Sloan Kettering Institute, New York, NY, USA., Rudensky AY; Howard Hughes Medical Institute and Immunology Program, Sloan Kettering Institute and Ludwig Center at Memorial Sloan Kettering Cancer Center, New York, NY, USA., Melnick AM; Division of Hematology and Medical Oncology, Department of Medicine, Weill Cornell Medical College, New York, NY, USA., Leslie CS; Computational and Systems Biology Program, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Jazyk:	angličtina
Zdroj:	BioRxiv : the preprint server for biology [bioRxiv] 2023 Jul 28. Date of Electronic Publication: 2023 Jul 28.
DOI:	10.1101/2023.07.27.550836
Abstrakt:	The identification of cell-type-specific 3D chromatin interactions between regulatory elements can help to decipher gene regulation and to interpret the function of disease-associated non-coding variants. However, current chromosome conformation capture (3C) technologies are unable to resolve interactions at this resolution when only small numbers of cells are available as input. We therefore present ChromaFold, a deep learning model that predicts 3D contact maps and regulatory interactions from single-cell ATAC sequencing (scATAC-seq) data alone. ChromaFold uses pseudobulk chromatin accessibility, co-accessibility profiles across metacells, and predicted CTCF motif tracks as input features and employs a lightweight architecture to enable training on standard GPUs. Once trained on paired scATAC-seq and Hi-C data in human cell lines and tissues, ChromaFold can accurately predict both the 3D contact map and peak-level interactions across diverse human and mouse test cell types. In benchmarking against a recent deep learning method that uses bulk ATAC-seq, DNA sequence, and CTCF ChIP-seq to make cell-type-specific predictions, ChromaFold yields superior prediction performance when including CTCF ChIP-seq data as an input and comparable performance without. Finally, fine-tuning ChromaFold on paired scATAC-seq and Hi-C in a complex tissue enables deconvolution of chromatin interactions across cell subpopulations. ChromaFold thus achieves state-of-the-art prediction of 3D contact maps and regulatory interactions using scATAC-seq alone as input data, enabling accurate inference of cell-type-specific interactions in settings where 3C-based assays are infeasible. Competing Interests: C.S.L. is an SAB member and co-inventor of IP with Episteme Prognostics, unrelated to the current study. M.G.K. is a SAB member of 858 Therapeutics and received honorarium from Kumquat, AstraZeneca and Consultancy with Transition Bio. A.D.V is an SAB member of Arima Genomics. A.Y.R. is an SAB member and has equity in Sonoma Biotherapeutics, Santa Ana Bio, RAPT Therapeutics and Vedanta Biosciences. He is an SEB member of Amgen and BioInvent and is a co-inventor or has IP licensed to Takeda that is unrelated to the content of the present study. A.M. has research funding from Janssen, Epizyme and Daiichi Sankyo. A.M. has consulted for Exo Therapeutics, Treeline Biosciences, Astra Zeneca. The remaining authors declare no competing interests.
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu