COCOA: A Framework for Fine-scale Mapping Cell-type-specific Chromatin Compartmentalization Using Epigenomic Information

kai li,ping zhang,jinsheng xu,Zi Wen,junying zhang,Zhike Zi,Li Li

DOI: https://doi.org/10.1101/2024.05.11.593669

2024-05-14

Abstract:Chromatin compartmentalization and epigenomic modification are crucial in cell differentiation and diseases development. However, precise mapping of chromatin compartmental patterns requires Hi-C or Micro-C data at high sequencing depth. Exploring the systematic relationship between epigenomic modifications and compartmental patterns remains challenging. To address these issues, we present COCOA, a deep neural network framework using COnvolution and attention mechanisms to infer fine-scale chromatin COmpartment pAtterns from six histone modification signals. COCOA extracts 1-D track features through bi-directional feature reconstruction after resolution-specific binning epigenomic signals. These track features are then cross-fused with contact features using an attention mechanism and transformed into chromatin compartment patterns through residual feature reduction. COCOA demonstrates accurate inference of chromatin compartmentalization at a fine-scale resolution and exhibits stable performance on test sets. Additionally, we explored the impact of histone modifications on chromatin compartmentalization prediction through in silico epigenomic perturbation experiments. Unlike obscure compartments observed with 1kb resolution high-depth experimental data, COCOA generates clear and detailed compartmental patterns, highlighting its superior performance. Finally, we demonstrated that COCOA enables cell-type-specific prediction of unrevealed chromatin compartment patterns in various biological processes, making it an effective tool for gaining chromatin compartmentalization insights from epigenomics in diverse biological scenarios.

Bioinformatics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to accurately map cell - type - specific chromatin compartmentalization patterns using epigenomic information. Specifically, chromatin compartmentalization and epigenomic modifications are key factors in cell differentiation and disease development. However, to map fine - grained chromatin compartment patterns across multiple cell types, Hi - C or Micro - C data with high sequencing depth are usually required. Exploring the systematic relationship between epigenomic modifications and compartment patterns remains a challenge. To address these challenges, the researchers proposed COCOA (A Framework for Fine - grained Mapping of Cell - type - Specific Chromatin Compartmentalization Using Epigenomic Information), a deep neural network - based method that infers reliable fine - grained chromatin compartment patterns from six representative histone modification signals through convolution and attention mechanisms.

The main contributions of COCOA include:
1. **Fine - grained Chromatin Compartmentalization Prediction**: COCOA can extract 1D trajectory features from epigenomic signals, cross - fuse with contact features after bidirectional feature reconstruction, and finally convert contact features into chromatin compartment patterns through residual feature reduction. This enables COCOA to accurately predict chromatin compartmentalization patterns at a fine - grained resolution.
2. **Impact of Epigenomic Modifications**: The researchers verified the impact of different histone modifications on chromatin compartmentalization through computational experiments. In particular, in high - depth experimental data with a 1kb resolution, COCOA can generate clear and detailed compartment patterns, while these details are difficult to observe in the experimental data.
3. **Cell - type - Specific Prediction**: COCOA can perform cell - type - specific predictions on unexposed chromatin compartment patterns in different biological processes, thereby providing insights into chromatin compartmentalization in a wide range of biological scenarios.

In conclusion, COCOA provides an effective method to understand chromatin compartmentalization through epigenomic information, overcoming the limitations of existing methods in terms of fine - grained resolution and cell - type - specificity.