Python软件包，用于与SRAdb进行交互并从SRA下载数据集。官方文档
命令行用法
pysradb支持命令行用法。文档正在进行中。有关一些快速使用说明，请参见cmdline。有关每个子命令的说明列表，请参见快速入门。
$ pysradbusage: pysradb [-h] [--version] [--citation]{metadb,metadata,download,search,gse-to-gsm,gse-to-srp,gsm-to-gse,gsm-to-srp,gsm-to-srr,gsm-to-srs,gsm-to-srx,srp-to-gse,srp-to-srr,srp-to-srs,srp-to-srx,srr-to-gsm,srr-to-srp,srr-to-srs,srr-to-srx,srs-to-gsm,srs-to-srx,srx-to-srp,srx-to-srr,srx-to-srs}...pysradb: Query NGS metadata and data from NCBI Sequence Read Archive.version: 0.9.0.Citation: 10.12688/f1000research.18676.1optional arguments:-h, --help show this help message and exit--version show program's version number and exit--citation how to citesubcommands:{metadb,metadata,download,search,gse-to-gsm,gse-to-srp,gsm-to-gse,gsm-to-srp,gsm-to-srr,gsm-to-srs,gsm-to-srx,srp-to-gse,srp-to-srr,srp-to-srs,srp-to-srx,srr-to-gsm,srr-to-srp,srr-to-srs,srr-to-srx,srs-to-gsm,srs-to-srx,srx-to-srp,srx-to-srr,srx-to-srs}metadb Download SRAmetadb.sqlitemetadata Fetch metadata for SRA project (SRPnnnn)download Download SRA project (SRPnnnn)search Search SRA for matching textgse-to-gsm Get GSM for a GSEgse-to-srp Get SRP for a GSEgsm-to-gse Get GSE for a GSMgsm-to-srp Get SRP for a GSMgsm-to-srr Get SRR for a GSMgsm-to-srs Get SRS for a GSMgsm-to-srx Get SRX for a GSMsrp-to-gse Get GSE for a SRPsrp-to-srr Get SRR for a SRPsrp-to-srs Get SRS for a SRPsrp-to-srx Get SRX for a SRPsrr-to-gsm Get GSM for a SRRsrr-to-srp Get SRP for a SRRsrr-to-srs Get SRS for a SRRsrr-to-srx Get SRX for a SRRsrs-to-gsm Get GSM for a SRSsrs-to-srx Get SRX for a SRSsrx-to-srp Get SRP for a SRXsrx-to-srr Get SRR for a SRXsrx-to-srs Get SRS for a SRX
Downloading SRAmetadb (optional)
pysradb可以利用SQLite数据库文件，该文件具有SRAdb项目提供的经过预处理的元数据。但是，对于0.9.5版，此数据库文件对任何操作都不是硬性要求。
SRAmetadb can be downloaded using:
wget -c https://starbuck1.s3.amazonaws.com/sradb/SRAmetadb.sqlite.gz && gunzip SRAmetadb.sqlite.gz
另外，您也可以使用pysradb下载它，默认情况下会将其下载到当前工作目录中：
pysradb metadb
使用pysradb
Mode: SRAmetadb or SRAWeb
pysradb的初始版本完全取决于SRAdb项目提供的SRAmnetadb.sqlite文件，我们将其称为SRAmetadb模式。但是，在`pysradb 0.9.5中，对SQLite文件的依赖已变为可选。在没有SQLite文件的情况下，操作使用usiNCBi的esrarch和esummary界面执行，该模式称为SRAweb模式。可以通过下载SQLite文件执行除搜索以外的所有操作。注意：SRAweb模式当前不完全支持附加标志，例如—desc，-detailed和-expand，将来的发行版中将支持这些标志。但是，在SRAweb和SRAmetadb模式下都可以使用将一个ID相互转换为另一个ID的所有基本功能。
Search
搜索所有包含“核糖体分析”的项目：
可以添加本地的数据库的命令。
pysradb search "ribosome profiling" | head
获取SRA元数据
pysradb metadata --db ./SRAmetadb.sqlite SRP000941 --assay --desc --expand | headstudy_accession experiment_accession sample_accession run_accession library_strategy batch biomaterial_provider biomaterial_type cell_type collection_method differentiation_method differentiation_stage disease donor_age donor_ethnicity donor_health_status donor_id donor_sex line lineage medium molecule passage sample_term_id sex source_name tissue tissue_depot tissue_typeSRP000941 SRX006235 SRS004118 SRR018454 ChIP-Seq NaN cellular dynamics international cell line NaN NaN none none none NaN NaN NaN NaN NaN h1 embryonic stem cell mteser genomic dna between 30 and 50 efo_0003042 male NaN NaN NaN NaNSRP000941 SRX006236 SRS004118 SRR018456 ChIP-Seq NaN cellular dynamics international cell line NaN NaN none none none NaN NaN NaN NaN NaN h1 embryonic stem cell mteser genomic dna between 30 and 50 efo_0003042 male NaN NaN NaN NaNSRP000941 SRX006237 SRS004118 SRR018455 ChIP-Seq NaN cellular dynamics international cell line NaN NaN none none none NaN NaN NaN NaN NaN h1 embryonic stem cell mteser genomic dna between 30 and 50 efo_0003042 male NaN NaN NaN NaNSRP000941 SRX006239 SRS004213 SRR019072 Bisulfite-Seq #2 thomson laboratory cell line NaN NaN na embryonic stem cell none NaN NaN NaN NaN NaN h1 na tesr genomic dna 27 efo_0003042 male NaN NaN NaN NaNSRP000941 SRX006239 SRS004213 SRR019080 Bisulfite-Seq #2 thomson laboratory cell line NaN NaN na embryonic stem cell none NaN NaN NaN NaN NaN h1 na tesr genomic dna 27 efo_0003042 male NaN NaN NaN NaNSRP000941 SRX006239 SRS004213 SRR019081 Bisulfite-Seq #2 thomson laboratory cell line NaN NaN na embryonic stem cell none NaN NaN NaN NaN NaN h1 na tesr genomic dna 27 efo_0003042 male NaN NaN NaN NaNSRP000941 SRX006239 SRS004213 SRR019082 Bisulfite-Seq #2 thomson laboratory cell line NaN NaN na embryonic stem cell none NaN NaN NaN NaN NaN h1 na tesr genomic dna 27 efo_0003042 male NaN NaN NaN NaNSRP000941 SRX006239 SRS004213 SRR019083 Bisulfite-Seq #2 thomson laboratory cell line NaN NaN na embryonic stem cell none NaN NaN NaN NaN NaN h1 na tesr genomic dna 27 efo_0003042 male NaN NaN NaN NaNSRP000941 SRX006239 SRS004213 SRR019084 Bisulfite-Seq #2 thomson laboratory cell line NaN NaN na embryonic stem cell none NaN NaN NaN NaN NaN h1 na tesr genomic dna 27 efo_0003042 male NaN NaN NaN NaN
将SRP转换为GSE
$ pysradb srp-to-gse SRP075720study_accession study_aliasSRP075720 GSE81903
下载整个项目
Pysradb 使从 SRA 下载数据集变得非常容易。
$ pysradb download --out-dir ./pysradb_downloads -p SRP063852
下载由 SRP/SRX/SRR 模仿 SRA 项目的层次结构组织。
仅下载某些感兴趣的样本
pysradb metadata SRP000941 --assay | grep 'study\|RNA-Seq' | pysradb download
如果可用，将使用aspera-client下载来自该项目的所有RNA-seq样本。另外，它也可以使用wget。
Python代码使用
用例1：获取元数据表（SRA运行表）
pysradb最简单的用例是，当您事先知道SRA项目ID（SRP）并只是想获取与其关联的元数据时。这通常反映在您从NCBI网站获得的SraRunTable.txt中。请参见SraRunTable的示例。
from pysradb import SRAdbdb = SRAdb('SRAmetadb.sqlite')df = db.sra_metadata('SRP098789')df.head()
用例2：明智地下载整个项目并安排实验
一旦获取了元数据并确保这是您正在寻找的项目，您将希望立即下载所有内容。 NCBI遵循以下层次结构：SRP => SRX => SRR。每个SRP（项目）具有多个SRX（实验），每个SRX内又具有多个SRR（运行）。我们想在我们的下载文件中模仿这个层次。这样做的原因很简单：在大多数情况下，您最关心SRX，并且希望以一种或另一种方式“合并”您的SRR。具有此层次结构可确保您的下游代码可以轻松处理此类情况，而无需担心需要合并哪些运行（SRR）。
from pysradb import SRAdbdb = SRAdb('SRAmetadb.sqlite')df = db.sra_metadata('SRP017942')db.download(df)
用例3：下载实验的子集
通常，您只需要处理项目（SRP）中的一小部分样本。考虑这个项目，该项目的数据涵盖了四种检测方法。
df = db.sra_metadata('SRP000941')print(df.library_strategy.unique())['ChIP-Seq' 'Bisulfite-Seq' 'RNA-Seq' 'WGS' 'OTHER']
但是，您可能只对分析RNA-seq样本感兴趣，而只想下载该子集。使用pysradb很简单，因为元数据可以是子集，就像您可以将pandas中的数据子集化一样。
df_rna = df[df.library_strategy == 'RNA-Seq']db.download(df=df_rna, out_dir='/pysradb_downloads')()
用例4：获取细胞类型/治疗信息表格样本属性
细胞类型/组织信息通常隐藏在sample_attributes列中，该列可以扩展：
from pysradb.filter_attrs import expand_sample_attribute_columnsdf = db.sra_metadata('SRP017942')expand_sample_attribute_columns(df).head()
用例5：搜索数据集
我们在SRA上执行的另一种常见操作是搜索，纯文本搜索。
如果要查找描述中在某处出现核糖体谱的所有项目：
df = db.search_sra(search_str='"ribosome profiling"')df.head()
若有收获，就点个赞吧
0 人点赞
