从同一行中提取以特定字符开头的字符串答案

【问题标题】：Extract strings starting with specific characters from the same line从同一行中提取以特定字符开头的字符串
【发布时间】：2017-07-11 13:46:02
【问题描述】：

首先抱歉，如果我在发布之前没有遇到类似问题的答案。我有一组（72）个基因注释文件。我想以如下格式提取 GO 术语（奖励将是其他注释术语）

HORVU1Hr1G002090           GO:0003824
HORVU1Hr1G002090           GO:0006527
HORVU1Hr1G002090           GO:0008295
HORVU1Hr1G002090           GO:0008792
HORVU1Hr1G005360           GO:0004497
HORVU1Hr1G005360           GO:0005506
HORVU1Hr1G005360           GO:0016705
HORVU1Hr1G005360           GO:0020037
HORVU1Hr1G005360           GO:0055114
HORVU1Hr1G087600           GO:0009055
HORVU1Hr1G087600           GO:0015035
HORVU1Hr1G087600           GO:0016705
.
.
.

我的 input_file 看起来像这样：

HORVU1Hr1G002090.11 HORVU1Hr1G002090    chr1H:4283580-4286133   HC_G    arginine decarboxylase 1    GO:0003824, GO:0006527, GO:0008295, GO:0008792  PF00278, PF02784    IPR000183, IPR002985, IPR009006, IPR022643, IPR022644, IPR022657, IPR029066 HORVU1Hr1G002090
HORVU1Hr1G005360.1  HORVU1Hr1G005360    chr1H:11579708-11582804 HC_G    Cytochrome P450 superfamily protein GO:0004497, GO:0005506, GO:0016705, GO:0020037, GO:0055114  PF00067 IPR001128, IPR002403, IPR017972    HORVU1Hr1G005360
HORVU1Hr1G087600.1  HORVU1Hr1G087600    chr1H:539679073-539680597   HC_G    Glutaredoxin family protein GO:0009055, GO:0015035, GO:0045454  PF00462 IPR002109, IPR011905, IPR012336 HORVU1Hr1G087600
HORVU1Hr1G087620.1  HORVU1Hr1G087620    chr1H:539699799-539703594   HC_G    S-adenosyl-L-methionine-dependent methyltransferases superfamily protein    none    PF10294 IPR019410, IPR029063 HORVU1Hr1G087620
HORVU1Hr1G089380.1  HORVU1Hr1G089380    chr1H:543801190-543806492   HC_G    Subtilisin-like protease    GO:0004252, GO:0006508  PF00082, PF05922    IPR000209, IPR010259, IPR015500, IPR023828 HORVU1Hr1G089380
HORVU1Hr1G093570.2  HORVU1Hr1G093570    chr1H:553490639-553492292   HC_G    Ribonuclease T2 family protein  GO:0003723, GO:0033897  PF00445 IPR001568 HORVU1Hr1G093570
HORVU1Hr1G093660.11 HORVU1Hr1G093660    chr1H:553651123-553709366   HC_G    ribonuclease 3  GO:0003723, GO:0033897  PF00445 IPR001568 HORVU1Hr1G093660
HORVU1Hr1G094970.1  HORVU1Hr1G094970    chr1H:556830249-556834411   HC_G    Mitochondrial outer membrane protein porin 5    none    none    IPR023614 HORVU1Hr1G094970
HORVU1Hr1G016140.3  HORVU1Hr1G016140    chr1H:49798715-49799683 HC_u    undescribed protein none    none    none HORVU1Hr1G016140

我很确定grep -oP 可以解决问题，或者更好的是awk 或sed。但我无法让它工作......请帮忙！

我尝试了以下方法，但没有成功：

grep -oP 'GO.*\[,|\|]\K+' input_file
grep -oP 'GO.*+' input_file #prints the whole line after the first match
sed -n 's/.*\(GO:[[:alnum:]]*\).*,/\1/p' input_file
grep -oP 'GO:.*' input_file | tr ',' '|'| tr ' ' '|' | tr '\t' '|'| awk 'BEGIN {FS=OFS="|"}; {for(i=10;i<NF;i++){if($i~/^GO/){a=$i}} print $(NF),a}'

提前非常感谢！

【问题讨论】：

标签： awk sed grep string-matching

【解决方案1】：

像这样尝试awk：

awk '{for(i=1;i<=NF;i++){if($i~"^GO:"){gsub(",","");print $2,$i}}}' input

它遍历每个字段以查找以GO: 开头的字段，如果找到，则打印字段$2，然后是匹配的GO: 字段。 gsub 用于删除尾随逗号。

【讨论】：