You just received a directory of paired-end FASTQ files. Before starting any analysis, it is worth spending a minute checking that they are complete and look as expected.

1. Check file integrity

seqfu check validates every file in a directory, pairing R1 and R2 automatically, and returns a non-zero exit status if something is wrong:

seqfu check --dir reads/

OK	PE	reads/sample1_R1.fq.gz	200	27553	0
OK	PE	reads/sample2_R1.fq.gz	200	36372	0
ERR	PE	reads/uneven_R1.fq.gz	-	900	3	Number of sequences in R1 and R2 do not match (7, 3);...

Here a truncated pair is caught before it can silently break a downstream tool.

2. Count the reads

seqfu count prints one line per sample, and checks that the two files of each pair contain the same number of reads:

seqfu count reads/*.fq.gz

3. Length statistics

seqfu stats reports lengths (and N50, which is more useful for assemblies). -n prints a table for the terminal, -b strips the paths:

seqfu stats -n -b reads/*_R1*.fq.gz

Add --multiqc stats_mqc.txt to include the numbers in a MultiQC report.

4. Quality encoding and profile

seqfu qual scans the first reads of each file, reports the possible encodings and the average quality:

seqfu qual reads/*.fq.gz

reads/sample1_R1.fq.gz	11.0	37.0	Sanger;Illumina-1.8;	36.27+/-3.50	87

5. Eyeball a few reads

Finally, look at some records with coloured quality bars, highlighting a primer if you expect one:

seqfu view -o CCTACGGGNGGCWGCAG reads/sample1_R1.fq.gz | less -R

For a full-screen, scrollable view use seqfu less.