Hey I am interested to know if this has been sorted out yet? As I am facing the same problem as well.

These are the commands that I used:-

soap index

2bwt-builder ref.fa

soap alignment

soap -a read1.fq -b read2.fq -D ref.index -o map.file -2 unmap.file -m 300 -x 500

Instead of using the map2sam.pl file, I used the soap2sam.pl file

sam2bam

samtools faidx ref.fa

samtools view -bt ref.fai -o read.bam read.sam

bam2sorted

samtools sort .bam .sorted

merge

samtools merge merge_read.bam bam1 bam2

pileup

samtools mpileup -ugf ref.fa merge_bam |bcftools view -bcvg -> merge_bcf

bcftools view merge_bcf >merge_vcf

This is one of the result from my flagstat file:- (my original file has 364617084 reads)

66478440 + 0 mapped (100%:nan %)

66478440 + 0 paired in sequencing

33239220 +0 read 1

33239220 +0 read 2

0 + 0 properly paired (0.00% :nan%)

66478440 + 0 with itself and mate mapped

0 + 0 singletons (0.00% :nan%)

0 + 0 with mate mapped to a different chr

0 + 0 with mate mapped to a different chr (mapQ>=5)

I'd love to hear some inputs. It seems like all of the mapping distributions for my paired-end test files on SOAP2 are 0% when it comes to being properly paired. Any idea what could go wrong?

Thanks a bunch!

Yintze

I have reviewed the sam file that converted from soap result and compare it to the sam file generated by bowtie2. I found that the result of soap have lost some information such as "AS, XN, XO, XG, etc.". Perhaps this is the reason.