DeLoop: a deep learning model for chromatin loop prediction from sparse ATAC-seq data
Luo, Y.; Zhang, Z.
Show abstract
Deciphering gene regulation and understanding the functional implications of disease-associated non-coding variants require the identification of cell-type-specific 3D chromatin interactions. Current chromosome conformation capture technologies fall short in resolution when handling limited cell inputs. To address this limitation, we introduce DeLoop, a deep learning model designed to predict CTCF-mediated chromatin loops from sparse ATAC-seq data by leveraging multitask learning techniques and attention mechanisms. Our model utilizes ATAC-seq data and DNA sequence features, showcasing superior performance compared to existing state-of-the-art models, particularly under low read depth conditions, enabling accurate chromatin loop inference when sufficient cells are infeasible. In addition, generalizing across cell types, DeLoop proves effective in de novo prediction tasks and its potential for predicting functional interactions.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
Similar papers in this journal
- NanoLoop: A deep learning framework leveraging Nanopore sequencing for chromatin loop prediction 97%
- DeDoc2 identifies and characterizes the hierarchy and dynamics of chromatin TAD-like domains in the single cells 96%
- Cross-species prediction of transcription factor binding by adversarial training of a novel nucleotide-level deep neural network 95%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.