1000g sequence reads PE or SE?

[ymc]

I have trouble understand which reads are single reads and which are pair ended reads

For example, there are three files in

ftp://ftp.1000genomes.ebi.ac.uk/vol1...sequence_read/

SRR035330_1.filt.fastq.gz
SRR035330_2.filt.fastq.gz
SRR035330.filt.fastq.gz

I checked the sequence index and they are all supposedly PAIR_ENDED. I think SRR035330_1.filt.fastq.gz and SRR035330_2.filt.fastq.gz constitute one pair. I think I can run them thru bwa to get two sai files and go from there.

But what is SRR035330.filt.fastq.gz??? Is it really pair ended reads? Or is it single reads?

Thanks a lot!

[adaptivegenome]

The third file is unpaired. I'm not sure which genomes you are using but many are sequenced using multiple platforms. For example, some are sequenced using SOLID and ILLUMINA.

[ymc]

Thanks for your reply. Suppose I want to generate a single bam file from them. Is this the right way to do this?

bwa aln -t 6 hg19.fa SRR035330.filt.fastq.gz > 0.sai
bwa aln -t 6 hg19.fa SRR035330_1.filt.fastq.gz > 1.sai
bwa aln -t 6 hg19.fa SRR035330_2.filt.fastq.gz > 2.sai
bwa samse hg19.fa 0.sai SRR035330.filt.fastq.gz | samtools view -bS - > se.bam
bwa sampe hg19.fa 1.sai 2.sai SRR035330_1.filt.fastq.gz SRR035330_2.filt.fastq.gz | samtools view -bS - > pe.bam
samtools merge SRR035330.bam se.bam pe.bam

[adaptivegenome]

That's the idea. And then of course sort and mark duplicates....

[ymc]

Ah. Now that I think about it. I think the SRR035330 run is a pair ended one. But for some reasons, only one end is sequenced for some fragments. This explains why the SRR035330.filt.fastq.gz is much smaller than the other two files.

Did I describe the situation correctly?

[adaptivegenome]

This is possible too. Sometimes you get unpaired reads from PE sequencing.