工具：如何在国内高速下载SRA

如何在国内高速下载SRA

下载公共测序数据，一般是通过在NCBI上搜索到study的Run SRA号，然后使用NCBI提供的prefetch下载数据，但因NCBI国内访问较为缓慢导致下载速度巨慢，因此推荐使用EBI提供的enaBrowserTools+Aspera的方式从ENA下载SRA。该下载方式相对aspera更好，只需要提供accession号即可（如SRR号），特推荐使用该方法下载。

###　软件

• Aspera

High-performace transfer brower plugin，IBM旗下的高速下载插件。

# 下载
wget wget https://download.asperasoft.com/download/sw/connect/3.9.9/ibm-aspera-connect-3.9.9.177872-linux-g2.12-64.tar.gz 
# 解压
tar -zxvf ibm-aspera-connect-3.9.9.177872-linux-g2.12-64.tar.gz
# 安装 软件安装路径是用户根目录的 .aspera/
./ibm-aspera-connect-3.9.9.177872-linux-g2.12-64.sh 

• enaBrowserTools

enaDataGet/enaGroupGet是enaBrowserTools的一个python脚本，enaBrowserTools是基于python3的与ENA web services接口的套件，可以方便访问ENA。安装方式简单，直接上github下载源代码，然后解压即可。

enaDataGet: all data for a given sequence, assembly, read or analysis accession or WGS set.

enaGroupGet: all data for a particular group(sequence, WGS, assembly, read or analysis) for a given sample or study accession.

# 下载&解压
wget　https://github.com/enasequence/enaBrowserTools/archive/v1.6.tar.gz
tar -zxvf v1.6.tar.gz

# 配置别名 方便查看脚本位置信息
alias enaDataGet=/disk/share/toolkits/enaBrowserTools/python3-1.6/enaDataGet
alias enaGroupGet=/disk/share/toolkits/enaBrowserTools/python3-1.6/enaGroupGet

配置

• 配置aspera：

# 配置许可(这一步需要账户有root权限，普通用户无法设置，也可以不需要设置)
sudo cp ~/.aspera/connect/etc/aspera-license /usr/local/bin/
# 添加环境变量
echo 'export PATH=~/.aspera/connect/bin:$PATH' >> ~/.bashrc
source ~/.bashrc 
#  配置秘钥
mkdir /home/zouhua/.aspera/config/ 
# 复制到配置目录
cp ~/.aspera/connect/etc/asperaweb_id_dsa.openssh /home/zouhua/.aspera/config/

• 配置enaDataGet

如果使用Aspera下载数据，则需要配置aspera_settings.ini文件

# step1 配置aspera_settings.ini
cd /disk/share/toolkits/enaBrowserTools-1.6 & vi aspera_settings.ini

配置前

[aspera]
ASPERA_BIN = /path/to/ascp
ASPERA_PRIVATE_KEY = /path/to/aspera_dsa.openssh
ASPERA_OPTIONS =
ASPERA_SPEED = 100M

配置后: 1. 指定ascp脚本；2.指定密钥；3.设置下载速度

[aspera]
ASPERA_BIN = /home/zouhua/.aspera/connect/bin/ascp
ASPERA_PRIVATE_KEY = /home/zouhua/.aspera/connect/etc/asperaweb_id_dsa.openssh
ASPERA_OPTIONS =
ASPERA_SPEED = 500M

# step2 配置aspera
export ENA_ASPERA_INIFILE="/disk/share/toolkits/enaBrowserTools-1.6/aspera_settings.ini"

查看帮助文档

• enaDataGet：重要参数 1. -f 指定数据类型；2. -d 指定本地下载目录；3. -a 指定是否使用aspera

usage: enaDataGet [-h] [-f {embl,fasta,submitted,fastq,sra}] [-d DEST] [-w]
                  [-m] [-i] [-a] [-as ASPERA_SETTINGS] [-v]
                  accession

Download data for a given accession

positional arguments:
  accession             Sequence, coding, assembly, run, experiment or
                        analysis accession or WGS prefix (LLLLVV) to download

optional arguments:
  -h, --help            show this help message and exit
  -f {embl,fasta,submitted,fastq,sra}, --format {embl,fasta,submitted,fastq,sra}
                        File format required. Format requested must be
                        permitted for data type selected. sequence, assembly
                        and wgs accessions: embl(default) and fasta formats.
                        read group: submitted, fastq and sra formats. analysis
                        group: submitted only.
  -d DEST, --dest DEST  Destination directory (default is current running
                        directory)
  -w, --wgs             Download WGS set for each assembly if available
                        (default is false)
  -e, --extract-wgs     Extract WGS scaffolds for each assembly if available
                        (default is false)
  -exp, --expanded      Expand CON scaffolds when downloading embl format
                        (default is false)
  -m, --meta            Download read or analysis XML in addition to data
                        files (default is false)
  -i, --index           Download CRAM index files with submitted CRAM files,
                        if any (default is false). This flag is ignored for
                        fastq and sra format options.
  -a, --aspera          Use the aspera command line client to download,
                        instead of FTP.
  -as ASPERA_SETTINGS, --aspera-settings ASPERA_SETTINGS
                        Use the provided settings file, will otherwise check
                        for environment variable or default settings file
                        location.
  -v, --version         show program's version number and exit
 
############################################################################
usage: enaGroupGet [-h] [-g {sequence,wgs,assembly,read,analysis}]
                   [-f {embl,fasta,submitted,fastq,sra}] [-d DEST] [-w] [-m]
                   [-i] [-a] [-as ASPERA_SETTINGS] [-t] [-v]
                   accession

Download data for a given study or sample, or (for sequence and assembly) taxon

positional arguments:
  accession             Study or sample accession or NCBI tax ID to fetch data
                        for

optional arguments:
  -h, --help            show this help message and exit
  -g {sequence,wgs,assembly,read,analysis}, --group {sequence,wgs,assembly,read,analysis}
                        Data group to be downloaded for this
                        study/sample/taxon (default is read)
  -f {embl,fasta,submitted,fastq,sra}, --format {embl,fasta,submitted,fastq,sra}
                        File format required. Format requested must be
                        permitted for data group selected. sequence, assembly
                        and wgs groups: embl and fasta formats. read group:
                        submitted, fastq and sra formats. analysis group:
                        submitted only.
  -d DEST, --dest DEST  Destination directory (default is current running
                        directory)
  -w, --wgs             Download WGS set for each assembly if available
                        (default is false)
  -e, --extract-wgs     Extract WGS scaffolds for each assembly if available
                        (default is false)
  -exp, --expanded      Expand CON scaffolds when downloading embl format
                        (default is false)
  -m, --meta            Download read or analysis XML in addition to data
                        files (default is false)
  -i, --index           Download CRAM index files with submitted CRAM files,
                        if any (default is false). This flag is ignored for
                        fastq and sra format options.
  -a, --aspera          Use the aspera command line client to download,
                        instead of FTP.
  -as ASPERA_SETTINGS, --aspera-settings ASPERA_SETTINGS
                        Use the provided settings file, will otherwise check
                        for environment variable or default settings file
                        location.
  -t, --subtree         Include subordinate taxa (taxon subtree) when querying
                        with NCBI tax ID (default is false)
  -v, --version         show program's version number and exit

使用

enaDataGet和enaGroupGet可配置aspera使用，参数为*-a/–aspera*，添加此参数则调用aspera。

• step1 先搜索accession在EBI的api接口；
• step2 本地创建下载日志文件目录 logs；
• step3 使用ascp软件下载accession；

/disk/share/toolkits/enaBrowserTools-1.6/python3/enaDataGet -f sra -a SRR212430 -d ./sra
# /disk/share/toolkits/enaBrowserTools-1.6/python3/enaDataGet -f sra SRR212430 -d ./sra

下载成功

下载失败: 出现session stop即为失败

• 完全下载失败

• 部分下载失败

不使用aspera下载

问题

1. 出现session stop信息，解决方案：

   • ASPERA_SPEED 设置在100 - 400 M之间；

   • root权限下开通 udp端口

     iptables -I INPUT -p udp --dport 3301 -j ACCEPT
     iptables -I OUTPUT -p udp --dport 3301 -j ACCEPT
     

参考

1. enaBrowserTools