You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
huangyechao 88cf3415ab README 图片修改 1 year ago
assets README 图片修改 1 year ago
tasks 第一次提交 1 year ago
README.md README 图片修改 1 year ago
inputs 调整markdown渲染 1 year ago
workflow.wdl 第一次提交 1 year ago

README.md

Author: Huang Yechao

E-mail:17210700095@fudan.edu.cn

Git: http://choppy.3steps.cn/huangyechao/target-somatic.git

Last Updates: 16/1/2019

Description

本 APP 所构建的是用于二代测序目标区域测序 somatic 分析流程。使用的软件是SentieonA fast and accurate solution to variant calling from next-generation sequence data 。本流程构建所使用的方法是基于流程语言WDL 并将其封装为Choppy平台上的APP进行使用。流程图如下所示:

somatic

  1. input:通常为二代目标区域测序所获得的fastq文件,通常包含 Tumornormal 两种类型的数据;此外还应当包含有测序时使用的 bed 文件
  2. Mapping:将测序所得的数据与参考基因组进行比对,找到每一条read在参考基因组上的位置,将结果信息储存在bam文件中,并对获得的 bam 文件进行质控
  3. Dedup:在制备文库的过程中,由于PCR扩增过程中会存在一些偏差,有的序列会被过量扩增。在比对的时候,这些过量扩增出来的完全相同的序列就会比对到基因组的相同位置。而这些过量扩增的reads并不是基因组自身固有序列,不能作为变异检测的证据,因此,要尽量去除这些由PCR扩增所形成的duplicates,并对去除重复之后的bam文件进行质控
  4. Realigner: 将比对到 indel 附近的 reads 进行局部重新比对,将比对的错误率降到最低
  5. BQSR:对bam文件里reads的碱基质量值进行重新校正,使最后输出的bam文件中reads中碱基的质量值能够更加接近真实的与参考基因组之间错配的概率
  6. co-realignment :将配对样本的 T/N 组合在一起
  7. TNseq :变异检测步骤,变异结果主要包括 SNVsINDELs
  8. TNscope :变异检测步骤,变异结果主要包括 SNVsINDELs以及 SV

App使用指南

安装App

# 激活choppy环境
source activate choppy-latest
# 安装app
choppy install huangyechao/target-somatic:<version>

准备samples文件

sample.csv 文件为提交任务时使用的输入文件,其内容是根据input文件中定义的信息对应生成的,也可使用 Choppysamples 功能生成:

choppy samples target-somatic --output samples.csv
#### samples.csv
normal_fastq_1,normal_fastq_2,tumor_fastq_1,tumor_fastq_2,regions,sample_name,cluster,disk_size,sample_id

其中sample_id对应于所分析样本的索引号,用于生成当前样本提交时的任务信息,应注意不要包含_,否则会出现报错。

提交任务

choppy batch target-germline samples.csv --project-name your_project

APP 构建

tasks

tasks目录中分析流程中每一个步骤的 WDL 文件,如 mapping.wdl 如下所示

task mapping {
String fasta
File ref_dir
File fastq_1
File fastq_2
   
String SENTIEON_INSTALL_DIR
String group
String sample
String pl
String docker
String cluster_config
String disk_size
   
command <<<
 set -o pipefail
 set -e	
 export SENTIEON_LICENSE=192.168.0.55:8990
 nt=$(nproc)
 ${SENTIEON_INSTALL_DIR}/bin/bwa mem -M -R "@RG\tID:${group}\tSM:${sample}\tPL:${pl}" -t $nt ${ref_dir}/${fasta} ${fastq_1} ${fastq_2} | ${SENTIEON_INSTALL_DIR}/bin/sentieon util sort -o ${sample}.sorted.bam -t $nt --sam2bam -i -
   	>>>
   
   runtime {
    dockerTag:docker
    cluster: cluster_config
    systemDisk: "cloud_ssd 40"
    dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
   }
   	output {
   		File sorted_bam = "${sample}.sorted.bam"
   		File sorted_bam_index = "${sample}.sorted.bam.bai"
   	}
   }
   

workflow

workflow.wdl 是定义了每一个步骤的输入文件以及各个步骤之间的以来关系的文件:

import "./tasks/mapping.wdl" as mapping
import "./tasks/Metrics.wdl" as Metrics
import "./tasks/Dedup.wdl" as Dedup
import "./tasks/deduped_Metrics.wdl" as deduped_Metrics
import "./tasks/Realigner.wdl" as Realigner
import "./tasks/BQSR.wdl" as BQSR
import "./tasks/corealigner.wdl" as corealigner
import "./tasks/TNseq.wdl" as TNseq
import "./tasks/TNscope.wdl" as TNscope
   
   
workflow {{ project_name }} {
   
File tumor_fastq_1
File tumor_fastq_2
File normal_fastq_1
File normal_fastq_2
   
String SENTIEON_INSTALL_DIR
String sample
String docker
   	
String fasta
File ref_dir
File dbmills_dir
String db_mills
File dbsnp_dir
File regions
String dbsnp
String disk_size
String cluster_config
   	
call mapping.mapping as tumor_mapping {
    input: 
   	group=sample + "tumor",
   	sample=sample + "tumor",
   	fastq_1=tumor_fastq_1,
   	fastq_2=tumor_fastq_2,
   	SENTIEON_INSTALL_DIR=SENTIEON_INSTALL_DIR,
   	pl="ILLUMINAL",
   	fasta=fasta,
   	ref_dir=ref_dir,
   	docker=docker,
   	disk_size=disk_size,
   	cluster_config=cluster_config
   }
   
call Metrics.Metrics as tumor_Metrics {
   	input:
   	SENTIEON_INSTALL_DIR=SENTIEON_INSTALL_DIR,
   	fasta=fasta,
   	ref_dir=ref_dir,
   	sorted_bam=tumor_mapping.sorted_bam,
   	sorted_bam_index=tumor_mapping.sorted_bam_index,		
   	sample=sample + "tumor",
   	docker=docker,
   	disk_size=disk_size,
   	cluster_config=cluster_config
   }
   
call Dedup.Dedup as tumor_Dedup {
   	input:
   	SENTIEON_INSTALL_DIR=SENTIEON_INSTALL_DIR,
   	sorted_bam=tumor_mapping.sorted_bam,
   	sorted_bam_index=tumor_mapping.sorted_bam_index,
   	sample=sample + "tumor",
   	docker=docker,
   	disk_size=disk_size,
   	cluster_config=cluster_config
   }
   	......
   	......
   }
   

其中文件最上面的 import 代表了所要使用的task文件,中间部分File/String xxx 表明了任务所传递出需要定义变量及其类型,call部分声明了流程的各个步骤及其依赖关系。(文档的具体说明详见WDL

input

input 文件为整个 APP 运行时所要输入的参数,对于可以固定的参数可以直接在input文件中给出,对于需要改变的参数用{{}}进行引用,将会使得参数在 samples 文件中出现;其中project_name为所运行的任务的名称,需要在提交任务是进行定义

{
"{{ project_name }}.fasta": "GRCh38.d1.vd1.fa",
"{{ project_name }}.ref_dir": "oss://pgx-reference-data/GRCh38.d1.vd1/",
"{{ project_name }}.dbsnp": "dbsnp_146.hg38.vcf",
"{{ project_name }}.dbsnp_dir": "oss://pgx-reference-data/GRCh38.d1.vd1/",
"{{ project_name }}.SENTIEON_INSTALL_DIR": "/opt/sentieon-genomics",
"{{ project_name }}.dbmills_dir": "oss://pgx-reference-data/GRCh38.d1.vd1/",
"{{ project_name }}.db_mills": "Mills_and_1000G_gold_standard.indels.hg38.vcf",
"{{ project_name }}.docker": "localhost:5000/sentieon-genomics:v2018.08.01 oss://pgx-docker-images/dockers",
"{{ project_name }}.sample": "{{ sample_name }}",
"{{ project_name }}.tumor_fastq_2": "{{ tumor_fastq_2 }}",
"{{ project_name }}.tumor_fastq_1": "{{ tumor_fastq_1 }}",
"{{ project_name }}.normal_fastq_1": "{{ normal_fastq_1 }}",
"{{ project_name }}.normal_fastq_2": "{{ normal_fastq_2 }}",
"{{ project_name }}.regions": "{{ regions }}",
"{{ project_name }}.disk_size": "{{ disk_size }}",
"{{ project_name }}.cluster_config": "{{ cluster if cluster != '' else 'OnDemand ecs.sn2ne.2xlarge img-ubuntu-vpc' }}"
      }

{{ cluster if cluster != '' else 'OnDemand ecs.sn2ne.2xlarge img-ubuntu-vpc' }}表示当没有指定cluster 的配置信息时,则默认使用 ecs.sn2ne.2xlarge

更多使用信息

详见Choppy使用说明