JOnTADS: a unified caller for TADs and stripes in Hi-C data
Zeng, Q.; Xiang, G.; Zhang, Y.; Hardison, R. C.; Li, Q.
Show abstract
Topologically associating domains (TADs) and stripes are important architectural structures on Hi-C data that are important for gene regulation. We present Joint Optimized nested TADs and Stripes (JOn-TADS), a unified caller for TADs and stripes in Hi-C data. JOnTADS effectively identifies hierarchical TADs and stripes in population Hi-C and micro-C datasets, and hierarchical TADs in single-cell Hi-C data. It provides robust identifications aligned with known biology and effectively captures interaction frequency variations in contact maps across diverse contexts. When multiple samples are available, JOn-TADS leverages shared information across samples in TAD identification, reducing unwanted variation in TAD boundary identification while preserving biological differences. This approach enables robust identifications in single-cell Hi-C data, effectively addressing challenges posed by data sparsity. JOnTADS is computationally efficient and requires minimal user tuning.
Matching journals
The top 3 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Simultaneous smoothing and detection of topological units of genome organization from sparse chromatin contact count matrices with matrix factorization 99%
- A unified encyclopedia of human functional DNA elements through fully automated annotation of 164 human cell types 96%
- Benchmarking algorithms for joint integration of unpaired and paired single-cell RNA-seq and ATAC-seq data 96%
Similar papers in this journal
Similar papers in this journal
Similar papers in this journal
- Prediction of single-cell chromatin compartments from single-cell chromosome structures by MaxComp 97%
- Epigenetics is all you need: A Transformer to decode chromatin structural compartments from the epigenome 97%
- HiCImpute: A Bayesian Hierarchical Model for Identifying Structural Zeros and Enhancing Single Cell Hi-C Data. 96%
"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.