Non-ATGC, small-case, 'N' characters in Fastq file

Question

[gauravdube007]

Hi All,

I am currently performing genome assembly. I have generated the consensus fastq file using the commands below. But the fastq file consists of lot non-ATGC characters (highlighted with bold). What are these characters and how to handle these?

Commands used to generate Fastq file:

bwa index ref.fa
bwa aln -t 9 ref.fa D2_R2.fastq -f D2_R2.sai && bwa aln -t 9 cocsa_ref.fa D2_R1.fastq -f D2_R1.sai
bwa sampe ref.fa D2_R1.sai D2_R2.sai D2_R1.fq D2_R2.fq > D2-aln-pe2.sam
samtools faidx ref.fa
samtools view -bt ref.fa.fai D2-aln-pe2.sam > D2-aln-pe2.bam
samtools sort D2-aln-pe2.bam D2-aln-pe2.bam.srt
samtools index D2-aln-pe2.bam.srt.bam
samtools mpileup -uf ref.fa D2-aln-pe2.bam.srt.bam | bcftools view -cg - | vcfutils.pl vcf2fq > CONSENSUS.fq

CONSENSUS.fq file looks like:

@scaffold_1
nnngtttggtggtagtattggtatttcaaacacgctaggtgtttgttggttttgagtagg
tgtagctggagtagactctatctccatttctctatcagtttgggcctctggccctaggct
ctcctgtctgttttcttgagtatttactacaatagtatcactgtctggcggcattttatt
actaagctcttttcttagtaagcaactagatggtctgtgtgtttttgttttcgtgagtga
gacgtgttcagattagctactttaccagcttctagctctatagcgcgtgggctgcacgag
ttggcactagttgtaatcgatttcttgggatggatttgtatataattcgctaaaattaca
cctattctgaaaaactcgnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
nnnnnnnnTAATGTTACAAGTAAYAAGAAGGATYCTYTCCTTRACAAATRACGAGATGGC

P.S: Please also convey, how to handle the small-case characters and 'N's ? Should we mask/remove them to get a better set of scaffolds?

Thanks in advance

Answer 1

[Brian Bushnell]

Lower case indicates masked sequences already (often due to low confidence); many tools will ignore them. I don't see any reason to remove them.

The non-ACGTN characters are IUPAC symbols typically indicating polymorphisms. I normally convert them to N before further processing.

[gauravdube007]

Thanks a lot Brian.