linux 中shell统计fasta文件中每条染色体中的碱基数目

 

1、测试数据

[root@centos7 test]# ls
test.fa
[root@centos7 test]# cat test.fa
>chr1
addgg
ddges
df
>chr2
ertfg
sdf
>chr3
edret
dfdff
sfdfd
d
>chr4
iejie
sdgg

 

2、在末尾生成一个标记染色体

[root@centos7 test]# sed -i '$a >chrxxx' test.fa
[root@centos7 test]# cat test.fa
>chr1
addgg
ddges
df
>chr2
ertfg
sdf
>chr3
edret
dfdff
sfdfd
d
>chr4
iejie
sdgg
>chrxxx

 

3、生成循环配置文件

[root@centos7 test]# grep ">chr" test.fa | sed '$d' | paste - <(grep ">chr" test.fa | sed '1d')
>chr1   >chr2
>chr2   >chr3
>chr3   >chr4
>chr4   >chrxxx
[root@centos7 test]# grep ">chr" test.fa | sed '$d' | paste - <(grep ">chr" test.fa | sed '1d') > conf.txt
[root@centos7 test]# ls
conf.txt  test.fa
[root@centos7 test]# cat conf.txt
>chr1   >chr2
>chr2   >chr3
>chr3   >chr4
>chr4   >chrxxx

 

4、统计每条染色体上的染色体数目

[root@centos7 test]# ls
conf.txt  test.fa
[root@centos7 test]# cat conf.txt
>chr1   >chr2
>chr2   >chr3
>chr3   >chr4
>chr4   >chrxxx
[root@centos7 test]# cat conf.txt | while read {i,j}; do sed -n "/$i/,/$j/{/$i\|$j/b; p}" test.fa | sed 's/[\t ]*//g' | paste -d "" -s | awk -v a=$i '{print a, length}' >> result.txt; done
[root@centos7 test]# ls
conf.txt  result.txt  test.fa
[root@centos7 test]# cat result.txt   ## 查看结果
>chr1 12
>chr2 8
>chr3 16
>chr4 9
[root@centos7 test]# cat test.fa      ## 验证结果
>chr1
addgg
ddges
df
>chr2
ertfg
sdf
>chr3
edret
dfdff
sfdfd
d
>chr4
iejie
sdgg
>chrxxx

 

5、或者

[root@centos7 test]# ls
test.fa
[root@centos7 test]# awk '/^>/{if (l!="") print l; print; l=0; next}{l+=length($0)}END{print l}' test.fa
>chr1
12
>chr2
8
>chr3
16
>chr4
9
[root@centos7 test]# cat test.fa   ## 验证结果
>chr1
addgg
ddges
df
>chr2
ertfg
sdf
>chr3
edret
dfdff
sfdfd
d
>chr4
iejie
sdgg