Juicer:Hi-C数据处理分析（一）

参考：

[1]. Durand, N.C., et al., Juicer Provides a One-Click System for Analyzing Loop-Resolution Hi-C Experiments. Cell Systems, 2016. 3(1): p. 95-98.
[2]. Ben Zouari, Y., et al., ChiCMaxima: a robust and simple pipeline for detection and visualization of chromatin looping in Capture Hi-C. Genome Biology, 2019. 20(1).
[3]. Forcato, M., et al., Comparison of computational methods for Hi-C data analysis. Nature Methods, 2017. 14(7): p. 679-685.
[4] HiC 主要分析内容
[5] How to analyze Hi-C data with Juicer and scaffold your genome assembly using 3D-DNA

介绍：

请移步微信公众号内容

从上图我们看到 Juicer 相比其他软件还是蛮有优势的。文章中介绍突出的两个优点是（1）流程化，使用简单。（2）高性能，平行计算，对大数据友好。

软件安装

主程序是github上的脚本，git clone 下就行，主要是依赖软件需要解决。官方要求的下面这些，我服务器上都有，所以很幸运。当然针对不同的服务器群并行运算有不同的脚本，这里主要用的是CPU这个目录下的脚本。

• For alignment and creation of the Hi-C pairs file merged_nodups.txt:
  • GNU CoreUtils
  • Burrows-Wheeler Aligner (BWA)
• For .hic file creation and Juicer tools analysis:
  • Java 1.7 or 1.8 JDK. (Alternative link for Ubuntu/LinuxMint). Minimum system requirements for running Java can be found at http://java.com/en/download/help/sysreq.xml
  • Latest Juicer Tools jar
• For peak calling:
  • CUDA and an NVIDIA GPU
  • The native libraries included with Juicer are compiled for CUDA 7. Other versions of CUDA can be used, but you will need to download the respective native libraries from JCuda.
  • For best performance, use a dedicated GPU. You may also be able to obtain access to GPU clusters through Amazon Web Services or a local research institution.

脚本原理（juicer.sh）

fastq等文件输入 ---》bwa比对 ---》排序 ---》合并 ---》去PCR重复 ---》生成 hic文件

代码流程

#创建工作目录，其后所有操作都在此目录下进行
mkdir ./opt 
cd opt
#克隆 juicer
git clone https://github.com/theaidenlab/juicer.git
ln -s juicer/CPU scripts
cd scripts/common
wget http://hicfiles.tc4ga.com.s3.amazonaws.com/public/juicer/juicer_tools.1.7.6_jcuda.0.8.jar
ln -s juicer_tools.1.7.6_jcuda.0.8.jar juicer_tools.jar
cd ../..
# 参考基因组建立索引，我用的是 Homo_sapiens_assembly38.fasta
mkdir references
cd references 
cp <path>/Homo_sapiens_assembly38.fasta  # 或者 ln -s <path>/Homo_sapiens_assembly38.fasta
bwa index  Homo_sapiens_assembly38.fasta
cd ..
# 添加限制性内切酶位点信息，注意自己的是不是 MboI 酶。
mkdir restriction_sites
cd restriction_sites/
python2 ../juicer/misc/generate_site_positions.py  MboI  hg38_MboI ../references/Homo_sapiens_assembly38.fasta # 生成了 hg38_MboI.txt 文件
awk 'BEGIN{OFS="\t"}{print $1, $NF}' hg38_MboI.txt > hg38.chrom.sizes
cd ..
# 添加 fastq 文件 (官方测试文件)
mkdir fastq && cd fastq
nohub wget http://juicerawsmirror.s3.amazonaws.com/opt/juicer/work/HIC003/fastq/HIC003_S2_L001_R1_001.fastq.gz &
nohub wget http://juicerawsmirror.s3.amazonaws.com/opt/juicer/work/HIC003/fastq/HIC003_S2_L001_R2_001.fastq.gz &
cd ..
# 运行 Juicer
bash scripts/juicer.sh  -d <path/opt>  -D <path/opt> -y restriction_sites/hg38_MboI.txt  -z references/Homo_sapiens_assembly38.fasta -p restriction_sites/hg38.chrom.sizes -s MboI 
# 上面-z -p 设置绝对路径吧，要不然会识别不了。我直接改软件源码了，-d -D 都设置了还需要绝对路径，作者写的不人性化啊

结果文件

结果文件都放在了生成的 aligned 中，主要文件是hic和stats。hic文件格式请看 https://github.com/theaidenlab/juicebox/blob/master/HiC_format_v8.docx

补充

Juicer 官网 http://aidenlab.org/documentation.html
Github https://github.com/aidenlab/juicer