SAMTool tview problem for paired-end sequences mapped by BFAST

[NanYu]

I mapped some 50bp paired-end solid sequence to my reference genome using BFAST. After I get the SAM file from the BFAST, I use SAMTool to create the sorted BAM file and use tview to look at the alignment.

For some reason, many pairs with both sequences mapped to the genome within the expected distance (1.5Kb) as shown in the SAM file (column 3 and 4), were shown with a "underline", which supposed to mean "Secondary or orphan".

Also, when I click "C" for cs view, and then all the 0-3 are white, is this normal?

Maybe I did something wrong during some of the steps?

Below are the major steps I did:

1.
solid2fastq -o Q20 Q20_F3.csfasta Q20_R3.csfasta Q20_F3_QV.qual Q20_R3_QV.qual
2.
bfast match -f ref.fa -A 1 -r Q20_paired.fastq >bfast.matches.Q20.bmf
3.
bfast localalign -f ref.fa -A 1 -x ColorMatrix.txt -m bfast.matches.Q20.bmf >bfast.aligned_Q20.baf
4.
bfast postprocess -f ref.fa -A 1 -U -x ColorMatrix.txt -O 1 -i bfast.aligned_Q20.baf >bfast_Q20.sam

I should not use "-U" for paired end in this step, but if I do not use "-U" and only use "-S 2000" (expect the insertion size), BFAST became very, very slow.

5.
samtools view -bhS bfast_Q20.sam > bfast_Q20.bam

6.
samtools sort bfast_Q20.bam bfast_Q20.sorted

7.
samtools index bfast_Q20.sorted.bam


and this is the command I use to view the alignment:
samtools tview bfast_Q20.sorted.bam ref.fa


Thanks for helping!

[nilshomer]

Could you give an example of one alignment in the SAM file? I would be happy to help.

[NanYu]

Thanks for your help!

Here is one example with one pair that looks OK to me:

1701_2041_1098 177 24 58305795 255 50M = 58304960 -835 TGNTTTGATCATTTGCATTACTGGGTTGTAGAAGAAAAAGATATGAGATC 33":_^\TFDJSWXTGO]WWWTX\__]_`````````````\^``````` PG:Z:bfast AS:i:2300 MQ:i:255 NM:i:1 NH:i:1 IH:i:1 HI:i:1 MD:Z:2T47 CS:Z:G02322213332200002202231101001213031310031232100..1 CQ:Z:?=:=<>A;:9>>@>;=<>><<;=9;;;874:4::,29682/,1:9<:!!3 CM:i:2 XA:i:3 XE:Z:-..-----------------------------------------------

1701_2041_1098 113 24 58304960 255 50M = 58305795 835 GAGGGAAAGCCTAGCTGACAGGACTTGCAGCTGAATTACGCGTGGAGAAG 6M."$2"4K"3MEQZQQXMUPQ^^SHSY]TU[[`XNR[_^[_```````^ PG:Z:bfast AS:i:1675 MQ:i:255 NM:i:1 NH:i:1 IH:i:1 HI:i:1 MD:Z:1T48 CS:Z:T22022201133313030212321310212021121232322321222.22 CQ:Z:;:=:=<<;=99<:814;>4>.=791.<9<,;13<,<53))9&(96!.6 CM:i:5 XA:i:3 XE:Z:--.2-21--2----------------------------------------


but in the SAMTool tview, it was underlined:

[NanYu]

Also, when display in the cs view ("-C"), it seems that the reverse strand is not properly displayed: all is shifted by one (to the left).


The first attached picture shows the nt view ("-N") and the 3rd and last row is lower case (meaning reverse).





The second picture shows when in cs view "-C", those two sequences were shifted (1 base).



Is this a known bug for SAMTools tview?

[nilshomer]

This would be a great question for the samtools help list. I am not sure how supported "tview" really is though.

[NanYu]

This would be a great question for the samtools help list. I am not sure how supported "tview" really is though.

Thanks!
I just posted this one to the samtools help list.