quality control from fastq to vcf

[dongshenglulv]

Hi all,

This is my framework to call SNP from 50 fastq files,
1> use bowtie to get SAM from fastq. reference genome: hg19.fa, paras in bowtie: -k 2 -v 2. Then I have 50 SAM files
2> convert SAM to BAM. cmd line: samtools view -bS s01.sam > s01.bam, then I have 50 BAM files
3> sort BAM. cmd line: samtools sort s01.bam > s01.sort.bam
4> convert 50 sorted BAM files to one vcf file.
cmd line:
samtools mpileup -P ILLUMINA -ugf hg19.fa *.sort.bam | bcftools view -bcvg - > var.raw.bcf
bcftools view var.raw.bcf | vcfutils.pl varFilter -D 2000 > var.flt.vcf

However, the quality of VCF file seems not good. I'm wondering that in which step can I do the quality control? thanks very much

[arcolombo698]

Initial FastQC

Hello.

Did you use trimmomatic to initially do QC on the original fastq. To my understanding the best results are gained by using QC on the fastq files initially.

[swbarnes2]

Your file is probably fine. That's just how mpileup works; it puts in pretty much EVERY discrepancy, so the ration of garbage to real discrepancies is very high. So filter your vcf based on depth of coverage, quality score, etc. You can eyeball individual SNPs in IGV, if you want.

[zillur]

Hi,

I am new in this arena. I am using bowtie.
1. I have aligned the paired-end reads included with bowtie2
bowtie2 -x lambda_virus -1 ../reads/reads_1.fq -2 ../reads/reads_2.fq -S eg2.sam

2. Converted the SAM file into BAM file
samtools view -bS eg2.sam > eg2.bam

3. Converted the BAM file to a sorted BAM file
samtools sort eg2.bam eg2.sorted

4. Generated variant calls in VCF format
samtools mpileup -uf lambda_virus.fa eg2.sorted.bam | bcftools view -O u - > eg2.raw.bcf

To view this format
bcftools view eg2.raw.bcf
By which I got a huge file mentioning SNPs like:
gi|9626243|ref|NC_001416.1| 48496 . G <X> 0 . DP=2;I16=0,2,0,0,65,2125,0,0,45,1773,0,0,28,634,0,0;QS=1,0;MQ0F=0 PL 0,6,37
gi|9626243|ref|NC_001416.1| 48497 . G <X> 0 . DP=2;I16=0,1,0,0,27,729,0,0,3,9,0,0,25,625,0,0;QS=1,0;MQ0F=0 PL 0,3,4
gi|9626243|ref|NC_001416.1| 48498 . T <X> 0 . DP=2;I16=0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0;QS=0,0;MQ0F=0 PL 0,0,0
gi|9626243|ref|NC_001416.1| 48499 . T <X> 0 . DP=2;I16=0,1,0,0,33,1089,0,0,42,1764,0,0,0,0,0,0;QS=1,0;MQ0F=0 PL 0,3,33
gi|9626243|ref|NC_001416.1| 48500 . A <X> 0 . DP=1;I16=0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0;QS=0,0;MQ0F=0 PL 0,0,0
gi|9626243|ref|NC_001416.1| 48501 . C <X> 0 . DP=1;I16=0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0;QS=0,0;MQ0F=0 PL 0,0,0
gi|9626243|ref|NC_001416.1| 48502 . G <X> 0 . DP=1;I16=0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0;QS=0,0;MQ0F=0 PL 0,0,0

so many of this. Now I need to asses the quality of the output. Can anyone give me some suggestions by which I can measure the quality of my data.

Best Regards
Zillur