这是一个统计fastq文件各类指标的工具。 这个工具命名为stat_fastq, 他可以定制输出的格式和指定输出指标。可以支持输入多个fastq文件，并且支持多线程执行.
软件的安装：
git clone https://github.com/thecgs/stat_fastq.git
bash ./stat_fastq/INSTALL.sh
stat_fastq文件是一个编译好的二进制文件，如果你需要重新编译，可以这样编译g++ stat_fastq.cpp -o stat_fastq，注意，编译完成的二进制文件需要和主程序stat_fastq.py在同一文件夹下
使用方法如下：
$ python stat_fastq.py -h
Usage:
example 1:
/nfs1/public2/User/chenguisen/01.biosoftware/stat_fastq/stat_fastq.py <fastq.fofn> [option]
example 2:
/nfs1/public2/User/chenguisen/01.biosoftware/stat_fastq/stat_fastq.py <file.fq.gz> [option]
example 3:
/nfs1/public2/User/chenguisen/01.biosoftware/stat_fastq/stat_fastq.py <file1.fq.gz> [file2.fq.gz] [file3.fq.gz] ... [option]
Option:
-o --output Save output file of tsv format [default: Close]
-t --Transposition Transposition Stdout format [default: Vertical]
-d --Distribution Open output "Reads length distribution" file [default: Close]
-g --ggplot Drawing Reads length distribution plot [default: Close]
-n --Reads_Num Close ouput Reads of Number [default: Open]
-b --Reads_Base Close ouput Reads of Bese Number [default: Open]
-10 --Q10 Close ouput Q10 [default: Open]
-20 --Q20 Close ouput Q20 [default: Open]
-30 --Q30 Close ouput Q30 [default: Open]
-40 --Q40 Close ouput Q40 [default: Open]
-50 --Q50 Close ouput Q50 [default: Open]
-qmi --Min_qual Close ouput Min quality value [default: Open]
-qma --Max_qual Close ouput Max quality value [default: Open]
-AT --AT_Bases Close ouput AT Beses of Number [default: Open]
-GC --GC_Bases Close ouput GC Beses of Number [default: Open]
-A --A_Bases Close ouput A Beses of Number [default: Open]
-T --T_Bases Close ouput T Beses of Number [default: Open]
-G --G_Bases Close ouput G Beses of Number [default: Open]
-C --C_Bases Close ouput C Beses of Number [default: Open]
-N --N_Bases Close ouput N Beses of Number [default: Open]
-lmi --Min_len Close ouput Read length Min value [default: Open]
-lma --Max_len Close ouput Read length Max value [default: Open]
-lme --Mean_len Close ouput Read length Mean value [default: Open]
-p --Phread_Type Close ouput Phread Type [default: Open]
-h --help Show the help message and exit
-v --version Show the version message
Note: The [option] can be anywhere
Datetime: 2022/10/12; Author: Guisen Chen; Email: [email protected]; Cite: https://github.com/thecgs/stat_fastq
例子一：
$ cat fastq.fofn
/data/2022/10-12/WT_1.fq.gz
/data/2022/10-12/M3_1.fq.gz
/data/2022/10-12/M3_2.fq.gz
/data/2022/10-12/WT_2.fq.gz
$ python stat_fastq.py fastq.fofn
WT_1 M3_1 M3_2 WT_2
Reads_Num 82561175 77216259 96803945 106000544
Reads_Base(nt) 12384176250 11582438850 14520591750 15900081600
Q10(%) 99.997% 99.9993% 99.9993% 99.9995%
Q20(%) 83.4571% 76.3126% 75.9509% 77.9955%
Q30(%) 72.6262% 64.9859% 64.3737% 65.6465%
Q40(%) 36.4314% 52.8567% 52.1012% 52.9124%
Min_qual 2 2 2 2
Max_qual 41 41 41 41
AT_Bases(%) 5105473290(41.2258%) 5676494935(49.0095%) 6982315472(48.0856%) 7493666735(47.1297%)
GC_Bases(%) 7278331828(58.7712%) 5905861065(50.9898%) 7538172361(51.9137%) 8406330226(52.8697%)
A_Bases(%) 2945803511(23.7868%) 3237282594(27.9499%) 4052007243(27.9052%) 4561738335(28.69%)
T_Bases(%) 2159669779(17.4389%) 2439212341(21.0596%) 2930308229(20.1804%) 2931928400(18.4397%)
G_Bases(%) 4815207048(38.8819%) 3452302746(29.8064%) 4462377776(30.7314%) 4906221582(30.8566%)
C_Bases(%) 2463124780(38.8819%) 2453558319(29.8064%) 3075794585(30.7314%) 3500108644(30.8566%)
N_Bases(%) 371132(0.00299682%) 82850(0.000715307%) 103917(0.000715653%) 84639(0.000532318%)
Min_len 150 150 150 150
Max_len 150 150 150 150
Mean_len 150 150 150 150
Phread_Type 33 33 33 33
例子二：
$ python stat_fastq.py fastq.fofn -t -A -T -G -C -10 -40
Reads_Num Reads_Base(nt) Q20(%) Q30(%) Min_qual Max_qual AT_Bases(%) GC_Bases(%) N_Bases(%) Min_len Max_len Mean_len Phread_Type
WT_1 82561175 12384176250 83.4571% 72.6262% 2 41 5105473290(41.2258%) 7278331828(58.7712%) 371132(0.00299682%) 150 150 150 33
M3_1 77216259 11582438850 76.3126% 64.9859% 2 41 5676494935(49.0095%) 5905861065(50.9898%) 82850(0.000715307%) 150 150 150 33
M3_2 96803945 14520591750 75.9509% 64.3737% 2 41 6982315472(48.0856%) 7538172361(51.9137%) 103917(0.000715653%) 150 150 150 33
WT_2 106000544 15900081600 77.9955% 65.6465% 2 41 7493666735(47.1297%) 8406330226(52.8697%) 84639(0.000532318%) 150 150 150 33
该项目的早期版本完全由python编写，起初，历遍fastq的代码如下：
import sys
import gzip
from itertools import islice
file = sys.argv[1]
def read_fastq(fastq:str):
n = 0
f = gzip.open(fastq, 'rb')
while True:
try:
name = next(islice(f,n,n+4,1)).strip().decode()
seq = next(islice(f,n,n+4,1)).strip().decode()
comment = next(islice(f,n,n+4,1)).strip().decode()
qual = next(islice(f,n,n+4,1)).strip().decode()
n += 4
yield (name, seq, puls, qual)
except StopIteration:
break
f.close()
for l in read_fastq(file):
print(l)
之后改用了第三方库pyfastx，快了很多，代码如下：
import pyfastx
with pyfastx.Fastq(fastq, build_index=False) as f:
for read in fq:
name, seq, quals = read
再后来搜到了一个C++版本（ https://github.com/haiwufan/fastq_stat ），根据这个版本修改了源码（增加了一些指标），速度大幅提升。并且根据python封装了这个工具
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent