HiCNN: a very deep convolutional neural network to better enhance the resolution of Hi-C data

Abstract

MotivationHigh-resolution Hi-C data are indispensable for the studies of three-dimensional (3D) genome organization at kilobase level. However, generating high-resolution Hi-C data (e.g. 5 kb) by conducting Hi-C experiments needs millions of mammalian cells, which may eventually generate billions of paired-end reads with a high sequencing cost. Therefore, it will be important and helpful if we can enhance the resolutions of Hi-C data by computational methods.ResultsWe developed a new computational method named HiCNN that used a 54-layer very deep convolutional neural network to enhance the resolutions of Hi-C data. The network contains both global and local residual learning with multiple speedup techniques included resulting in fast convergence. We used mean squared errors and Pearson’s correlation coefficients between real high-resolution and computationally predicted high-resolution Hi-C data to evaluate the method. The evaluation results show that HiCNN consistently outperforms HiCPlus, the only existing tool in the literature, when training and testing data are extracted from the same cell type (i.e. GM12878) and from two different cell types in the same or different species (i.e. GM12878 as training with K562 as testing, and GM12878 as training with CH12-LX as testing). We further found that the HiCNN-enhanced high-resolution Hi-C data are more consistent with real experimental high-resolution Hi-C data than HiCPlus-enhanced data in terms of indicating statistically significant interactions. Moreover, HiCNN can efficiently enhance low-resolution Hi-C data, which eventually helps recover two chromatin loops that were confirmed by 3D-FISH.Availability and implementationHiCNN is freely available at http://dna.cs.miami.edu/HiCNN/.Supplementary information Supplementary data are available at Bioinformatics online.

DeepHiC: A generative adversarial network for enhancing Hi-C data resolution

“…For baseline models, we only performed comparisons on data downsampled to 1/16 reads as they commonly used in their study [17,25,26]. The python source code for HiCPlus was obtained from https://github.com/zhangyan32/HiCPlus_pytorch, together with the codes for data processing and pre-trained model parameter file.…”

“…Carron et al proposed a computational method called Boost-HiC for boosting reads counts of long-range contacts [25]. And Liu et al proposed HiCNN [26] which is a 54-layer CNN and achieved better performance than HiCPlus. While these results were encouraging, three problems still exist in Hi-C data resolution enhancement algorithms.…”

DeepHiC: A Generative Adversarial Network for Enhancing Hi-C Data Resolution

“…For baseline models, we only performed comparisons on data downsampled to 1/16 read as they commonly used in their study [17,25,26]. The python source code for HiCPlus was obtained from https://github.com/zhangyan32/HiCPlus_pytorch, together with the codes for data processing and pretrained model parameter file.…”

SRHiC: A Deep Learning Model to Enhance the Resolution of Hi-C Data

“…HiCPlus (Zhang et al, 2018) is the first work that applies a CNN, in which network architecture is similar to SRCNN (Dong et al, 2014), to enhance the resolution of Hi-C data. HiCNN (Liu and Wang, 2019), which was based on DRRN (Tai et al, 2017), used a very deep convolutional neural with 54 layers to predict the high-resolution Hi-C contact matrix from the low-resolution Hi-C contact matrix. HiCNN showed better prediction accuracy than HiCPlus, but the computational cost of HiCNN is much higher than that of HiCPlus.…”