ID转换专题之linux命令处理1：gtf

• ID转换的目的是寻找两列ID的对应关系，一列是当前数据你所有的，另一列ID是所需要的ID即可，不多不少，就两列即可。那么linux命令也是可以处理的，找一个这些信息的文本，用linux命令处理好。前提是linux基础命令你掌握了，掌握要求，培训课的第二天所讲内容。
• 下面我处理的是ENSG转gene name的gff文件，因为这个文件里的信息含义这两列，我只需提取出来就好了。

下载gtf

你可以用gencode的gtf或者ncbi的gff，里面含有你想要的ID信息就好，只是处理的代码不同，因此linux基础和进阶命令都要学会，很有用，很高效

ENSG ID与gene nama对应关系

$ zless -S gencode.v31.annotation.gtf.gz|grep -w 'gene'|cut -f 9|awk -v OFS="\t" '{print $2,$6,$4}'|sed 's/[";]//g'|sed '1i #gene_id\tgene_name\tgene_type'|less
# 或者用下面awk的gsub来做
zless -S gencode.v31.annotation.gtf.gz|grep -w 'gene'|cut -f 9|awk -v OFS="\t" '{gsub(/[";]/,"");print $2,$6,$4}'|sed '1i #gene_id\tgene_name\tgene_type'|less

image.png

ENST ID与ENSG ID对应关系

$ zless -S gencode.v31.annotation.gtf.gz|grep -w 'transcript'|cut -f 9|awk -v OFS="\t" 'BEGIN{print "gene_id","transcript_id","gene_name","transcript_type"}{print $2,$4,$8,$10}'|sed 's/[";]//g'|less -S
# zless -S gencode.v31.annotation.gtf.gz|grep -w 'transcript'|cut -f 9|awk -v OFS="\t" 'BEGIN{print "gene_id","transcript_id","gene_name","transcript_type"}{print $2,$4,$8,$10}'|sed 's/[";]//g'|sort|uniq|less -S
# gene_id transcript_id   gene_name       transcript_type4列信息

示例

如果你需要过滤类型查看的话

$ zless -S gencode.v31.annotation.gtf.gz|grep -w 'transcript'|awk '{print $18}'|sort|uniq
# 查看所有类型
"IG_C_gene";
"IG_C_pseudogene";
"IG_D_gene";
"IG_J_gene";
"IG_J_pseudogene";
"IG_pseudogene";
"IG_V_gene";
"IG_V_pseudogene";
"lncRNA";
"miRNA";
"misc_RNA";
"Mt_rRNA";
"Mt_tRNA";
"nonsense_mediated_decay";
"non_stop_decay";
"polymorphic_pseudogene";
"processed_pseudogene";
"protein_coding";
"pseudogene";
"retained_intron";
"ribozyme";
"rRNA";
"rRNA_pseudogene";
"scaRNA";
"scRNA";
"snoRNA";
"snRNA";
"sRNA";
"TEC";
"transcribed_processed_pseudogene";
"transcribed_unitary_pseudogene";
"transcribed_unprocessed_pseudogene";
"translated_processed_pseudogene";
"translated_unprocessed_pseudogene";
"TR_C_gene";
"TR_D_gene";
"TR_J_gene";
"TR_J_pseudogene";
"TR_V_gene";
"TR_V_pseudogene";
"unitary_pseudogene";
"unprocessed_pseudogene";
"vaultRNA";


# 然后用grep过滤上面的结果文件即可