PGC (Population genetic analysis of SVs in Chinese)

Description

Structural variants in Chinese population and their impact on phenotypes, diseases and population adaptation

(Characterization of structural variants of Chinese population by long-read sequencing)

In order to make sure the results are reproduceable, the pipeline is performed using framework Snakemake coupled with the environment conducted by Anoconda. And the pipeline can be used in other population with long-read sequencing.

The pipeline mainly contains:

• Characterization of structural variants and population genetic analysis

Requirements

Softwares:

• bedtools=2.27.1
• biopython=1.73
• minimap2=2.15
• mosdepth=0.2.5
• nanofilt=2.2.0
• nanoplot=1.20.0
• nanoqc=0.8.1
• nanovar=1.3.6
• nanosv==1.2.4
• samtools=1.9
• seqkit=0.10.1
• snakemake=5.4.2
• sniffles=1.0.10
• ucsc-liftover=377

Python packages:

• biopython=1.73
• gseapy=0.9.15
• pysam=0.15.2
• tinyfasta=0.1.0
• networkx=2.3
• numpy=1.16.4

R packages:

• r-argparser=0.4
• r-base=3.5.1
• r-dplyr=0.7.8
• r-futile.logger=1.4.3
• r-ggplot2=2.2.1
• r-ggpubr=0.2.4
• r-gridextra=2.3
• r-qqman=0.1.4
• r-readr=1.3.1
• r-reshape2=1.4.3
• r-scales=1.0.0
• r-scrime=1.3.5
• r-venndiagram=1.6.20

Other packages should be manually installed:

• EIGENSOFT=v7.2.1
• IGV=v2.8.6

The softweres and packages with corresponding versions were tested and used in our pipeline and manuascript. You can install the packages with different versions if you make sure these packages can run.

Configure the environment

manually install packages

• First install Annoconda.

For instance:

wget https://repo.anaconda.com/archive/Anaconda3-2020.07-Linux-x86_64.sh
bash Anaconda3-2020.07-Linux-x86_64.sh

• You can install all softwares and packages in one anaconda environment, or in multiple environments. If you change to one new environment, you can type this common:

source activate new-environment

• Creat the environment and install the packages

conda create -n NanoSV

• Most packages can be found in anaconda. For instance, bedtools, you can install like this:

conda install -c bioconda bedtools

create environment based on config file

Create the environment and install the packages using the config file:

conda env create -f environment.yml

Quick start for the pipeline

Just need clone and copy this package to your destination path.

In the directory pipeline of the package, there are pipeline file Population.pipeline.py and the config file Population.pipeline.yaml.

For example, we install all packages in the environment with name NanoSV, then we can run the pipeline:

source activate NanoSV
snakemake -p  -s ~/github/NanoHub/pipeline/Population.pipeline.py --configfile ~/github/NanoHub/pipeline/Population.pipeline.yaml -j 24

Introduction of pipeline

Config file

The config file of parameters Population.pipeline.yaml

For instance:

### Parameters
CondaENV: /home/wuzhikun/anaconda3/envs/NanoSV
PIPENV: /home/wuzhikun/github/PGC
ProjectPath: /home/wuzhikun/Project/Population

### target samples
SAMPLES:
  - CN196
  - CN068

First three parameters are most import.

• CondaENV indicates the anoconda environment which contain installed softwares and packages.
• PIPENV indicates the path of this pipeline.
• ProjectPath indicates the path of project which contain the data for analysis.

Pipeline file

The pipeline file Population.pipeline.py

For example:

### config parameter
CondaENV = config["CondaENV"]

### rule of pipeline
include: RULE_DIR + '/BaseFunction.rule.py'

### outcomes of pipeline
rule all:
    input:
        IN_PATH + "/QualityControl/Samples_quality_summary.xls",

The pipeline file mainly contain three parts:

• Parameter from .yaml config file
• Rule used in pipeline
• Control the pipeline and get the results

Rule files

The rule file is based on the function module:

• databaseFormat.rule.py: Change format of compared datasets
• BaseFunction.rule.py: Base function of python
• PopQuality.rule.py: Quality control for nanopore reads
• NanoCallSV.rule.py: Call SVs based on BAM file using multiple tools
• PopMultipleSV.rule.py: SV calling from multiple callers for each sample
• PopSVFilt.rule.py: High-confidence SVs after three filtering steps
• PopFrequency.rule.py: Allele frequency analysis for non-redundant SVs of the population
• populationGeno.rule.py: Genotypes of SVs in population
• PopStatistics.rule.py: Statistics for non-redundant SVs
• PopStructure.rule.py: Population structure analysis based on SVs
• PopAnnotation.rule.py: Annotation for SVs relative to genomic location
• PopOverlap.rule.py: Comparison to other published SV dataasets
• PopMergeSeq.rule.py: Analysis of SVs associated sequences

Tests

The pipeline was tested based on different rules, and the log files were in tests directory of the package.

• SV_Annotation-1.log
• SV_Annotation-2.log
• SV_Diversity.log
• SV_Filt.log
• SV_Genotype.log
• SV_Heter.log
• SV_Overlap.log
• SV_QualityControl.log
• SV_Statistics.log

Data

• Raw data: HRA000792

• VCF file: GVM000132

Contributions:

• Zhikun Wu: [email protected]
• Tong Li: [email protected]
• Zehang Jiang: [email protected]

Cite:

Wu, Z., Jiang, Z., Li, T. et al. Structural variants in the Chinese population and their impact on phenotypes, diseases and population adaptation. Nat Commun 12, 6501 (2021). https://doi.org/10.1038/s41467-021-26856-x