HIC simple process
1，什么是Hic数据？
Hi-C是研究染色质三维结构的一种方法。Hi-C技术源于染色体构象捕获（Chromosome Conformation Capture, 3C）技术，利用高通量测序技术，结合生物信息分析方法，研究全基因组范围内整个染色质DNA在空间位置上的关系，获得高分辨率的染色质三维结构信息。
2，Hic数据的优势
- 通过Scaffold间的交互频率大小，可以对已组装的基因组序列进行纠错。
- 基因信息不再仅仅是contig片段，而是被划分至染色体上，成为染色体水平。
- 无需辛苦的构建群体，单一一个体就能实现染色体定位。
- 相比遗传图谱，标记密度更大，序列定位更完整。
- 可以开展染色体重排等结构变异研究。
- QTL、GWAS可以定位区间到某个染色体。
- 可以解析该物种的三维基因结构、染色体互作及动态变化。
3，目前的处理流程
4，分析主要工具
目前针对Hi-c数据处理的工具主要是Hic-pro和juicer
#####HIC图谱，TAD结构，loop结构，3D-建模
####HiC-Pro installlation####
wget -c http://github.com/nservant/HiC-Pro/archive/refs/tags/v3.1.0.tar.gz
tar -zxvf HiC-Pro-3.1.0.tar.gz
conda env create -f /data5/tan/zengchuanj/Software/HiC-Pro-3.1.0/environment.yml -p /data5/tan/zengchuanj/conda/conda/envs/HiC-Pro
conda activate HiC-Pro
#configure.install.txt:
PREFIX = /data5/tan/zengchuanj/Software/HiC-Pro-3.1.0
BOWTIE2_PATH = /data5/tan/zengchuanj/conda/conda/envs/HiC-Pro/bin/bowtie2
SAMTOOLS_PATH = /data5/tan/zengchuanj/conda/conda/envs/HiC-Pro/bin/samtools
R_PATH = /data5/tan/zengchuanj/conda/conda/envs/HiC-Pro/bin/R
PYTHON_PATH = /data5/tan/zengchuanj/conda/conda/envs/HiC-Pro/bin/python
CLUSTER_SYS = TORQUE
make configure
make install
ref_dir = /data5/tan/zengchuanj/pipeline/Annotation/HIC/GRCm39.genome.fa.gz
gunzip GRCm39.genome.fa.gz
#build index
pwd:/data5/tan/zengchuanj/pipeline/Annotation/HIC
bowtie2-build GRCm39.genome.fa mouse
samtools faidx GRCm39.genome.fa
#基因组中序列大小文件
awk '{print $1 "\t" $2}' GRCm39.genome.fa.fai > mouse.genome.sizes
#创建酶切位点文件
bin=/data5/tan/zengchuanj/Software/HiC-Pro-3.1.0/bin/utils/digest_genome.py
#python $bin GRCm39.genome.fa -r mobi -o mouse_mobi.bed
python $bin GRCm39.genome.fa -r ^GATCGATC -o mouse_mobi.bed
#config-hicpro.txt:
N_CPU，CPU数目；
BOWTIE2_IDX_PATH，索引所在目录
REFERENCE_GENOME，比对参考基因组路径及前缀
GENOME_SIZE，chrom.sizes文件的路径
GENOME_FRAGMENT，酶切片段的bed文件的路径
LIGATION_SITE，酶切位点末端补平再次连接后形成的嵌合序列，例如HindIII，则为AAGCTAGCTT；如果是MboI则序列为GATCGATC；
## SYSTEM AND SCHEDULER - Start Editing Here !!
N_CPU = 50 #CPU线程数
LOGFILE = hicpro.log #log文件名
JOB_NAME = hicpro #任务名
JOB_MEM = 100gb #占用内存
JOB_WALLTIME =
JOB_QUEUE =
JOB_MAIL =
PAIR1_EXT = _R1
PAIR2_EXT = _R2
BOWTIE2_IDX_PATH = /data5/tan/lishix/jys/test/results/reads #比对的reads文件目录
BOWTIE2_GLOBAL_OPTIONS = --very-sensitive -L 30 --score-min L,-0.6,-0.2 --end-to-end --reorder
BOWTIE2_LOCAL_OPTIONS = --very-sensitive -L 20 --score-min L,-0.6,-0.2 --end-to-end --reorder
GENOME_SIZE = /data5/tan/zengchuanj/pipeline/Annotation/HIC/mouse.genome.sizes #genome.sizes的绝对路径
## Digestion Hi-C
GENOME_FRAGMENT = /data5/tan/zengchuanj/pipeline/HIC/mouse_mobi.bed #绝对路径
LIGATION_SITE = GATCGATC #限制性内切酶，具体用的什么酶可以咨询测序公司，我这里用的Mboi
MIN_FRAG_SIZE = 100
MAX_FRAG_SIZE = 100000
MIN_INSERT_SIZE = 100
MAX_INSERT_SIZE = 1000
## Contact Maps
BIN_SIZE = 20000 40000 150000 500000 1000000 #根据自身需求设置 bin size
MATRIX_FORMAT = upper
/data5/tan/zengchuanj/Software/HiC-Pro-3.1.0/bin/HiC-Pro -c /data5/tan/zengchuanj/pipeline/HIC/HiC-Pro/config-hicpro.txt -i /data5/tan/zengchuanj/pipeline/HIC/HiC-Pro/fastq -o /data5/tan/zengchuanj/pipeline/HIC/HiC-Pro/results
#目录构成：
fastq/sample:
sample_R1.fastq.gz
sample_R2.fastq.gz
#####juicer installation####
conda create -n juicer -c bioconda bwa -y
conda activate jucier
mkdir work && mkdir references && mkdir restriction_sites
Juicer/juicer/references # 存放参考基因组相关文件的文件夹
Juicer/juicer/work # 存放样本的序列文件，和分析结果的文件夹
Juicer/juicer/restriction_sites # 存放参考基因组酶切图谱的文件夹
wget https://github.com/aidenlab/juicer/archive/refs/tags/1.6.tar.gz
tar -xzvf juicer-1.6.tar.gz
ln -s juicer/CPU scripts
# scripts 应该在juicer目录下
cd juicer/scripts/common
wget -c https://hicfiles.tc4ga.com/public/juicer/juicer_tools.1.9.9_jcuda.0.8.jar
ln -s juicer_tools.1.9.9_jcuda.0.8.jar juicer_tools.jar
#构建基因组索引
pwd:/data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/references
bwa index GRCm39.genome.fa
#生成酶切图谱文件
python /data5/tan/zengchuanj/Software/juicer/misc/generate_site_positions.py Mboi genome /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/references/GRCm39.genome.fa
#生成染色体长度文件
# genome_DpnII.txt 文件由上一步生成
awk 'BEGIN{OFS="\t"}{print $1, $NF}' genome_Mboi.txt > genome.chrom.sizes
cd ./references
python /data5/tan/zengchaunj/pipeline/HIC/Juicer/misc/generate_site_positions.py Mboi mm9 mm9.fasta
# 三个参数分别为 内切酶名称，参考基因组名称，参考基因组序列文件的路径
nohup bash scripts/juicer.sh -d /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/test -D /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer -y /data5/tan/lishix/HIC/opt/juicer/restriction_sites/mm39_MboI.txt -z /data5/tan/lishix/HIC/opt/juicer/references/Mus_musculus.GRCm39.dna.toplevel.fa -p restriction_sites/genome.chrom.sizes -s MboI -t 10 2> test.txt &
Usage:
# nohup 命令会将程序挂在后台运行
nohup bash /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/scripts/juicer.sh \
-z /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/references/GRCm39.genome.fa \
-p /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/restriction_sites/genome.chrom.sizes \
-y /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/restriction_sites/GRCm39.genome_MboI.txt \
-s MboI \
-d /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/work/ \
-D /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer \
-t 40 > log.txt &
# -z参数指定参考基因组fasta所在路径，在该路径下必须同时存在对应的bwa索引
# -p参数指定染色体长度文件；
# -y指定基因组酶切图谱的路径；
# -d指定样本原始文件存放的路径；
# -D指定软件的安装路径，
# -t指定bwa比对使用的线程数，默认是使用全部线程。
#HIC图谱绘制
data_dir = /data5/tan/lishix/jys/test/results/
species = mouse
酶：mboi
#使用HiCPlotter.py对HiC-Pro结果进行可视化
python2.7 HiCPlotter.py -o genome \
-f genome_500000_iced.matrix \
-r 500000 -tri 1 \
-bed genome_500000_abs.bed \
-n genome \
-wg 1 -chr chromosome7
-o 输出的文件名
-f _500000_iced.matrix产生的矩阵文件
-r 矩阵的分辨率
-bed _500000_abs.bed产生的bed文件
-n 输出图片最上方的名字
-chr 最后一号染色体的名字 可使用"tail -n 1 *.bed"命令查看
#使用juicer call tad
ref:https://github.com/aidenlab/juicer/wiki/Arrowhead
/data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/scripts/common/juicer_tools arrowhead --ignore_sparsity /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/work/aligned/inter.hic ./contact_domains_list/
##使用juicer call loop
nohup java -jar /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/scripts/common/juicer_tools.jar hiccups --cpu --threads 19 -r 5000,10000 --ignore_sparsity /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/work/aligned/inter.hic inter.hic.hiccups > loop.txt &
nohup java -jar /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/scripts/common/juicer_tools.jar hiccups --gpu --threads 19 -r 2500,5000,7500,10000,12500,15000,17500,20000,22500 --ignore_sparsity /data5/tan/zengchuanj/pipeline/HIC/Juicer/juicer/work/aligned/inter.hic inter.hic.hiccups > loop.txt &
