Mate reads to fastq

[Tomi]

Hi,

I am trying to extract mate reads out of a sam file with following flags: view -b -S -f 8 -F 4 output.sam > mate.bam

and then with bam2fastq: bam2fastq -o mate#.fastq -f mate.bam

Unfortunetly i get this error message:


This looks like paired data from lane 0.
Output will be in unmapped_1.fastq and unmapped_2.fastq
1 sequences in the BAM file
1 sequences exported
WARNING: 1 reads could not be matched to a mate and were not exported


Probably someone can help me out?

Thanks,
TOmi

[maubp]

Is there really just one read in the original SAM file, and therefore just 1 in the BAM file?

[Tomi]

Hi,

thank you for your reply.
No both reads are in the sam file (reverse and forward).

Greetings,
Tomi

[maubp]

Could you share the (start of the) SAM file then? If you post it here, use the [ code ] tags - e.g. via the # icon on the advanced editor.

[Tomi]

BWA Sampe Output

@SQ     SN:gi|157704448|ref|AC_000133.1|        LN:219475005
@PG     ID:bwa  PN:bwa  VN:0.5.9-r16
testSample_0_1  73      gi|157704448|ref|AC_000133.1|   1       37      75M     =       1       0       ATTGACAAGGGGAGGGAAAAGAGGAACAGAAATTCTTTTCTAT$
testSample_0_2  133     gi|157704448|ref|AC_000133.1|   1       0       *       =       1       0       ATACCCAGGATTTTACCTGTAAAAGTACCCTCAGGTCGTGATT$
testSample_1_1  77      *       0       0       *       *       0       0       ATCGTCAATAGGGTACTACTTCCATAATTTTGTAAATCCGCATGTTCCTCGAATAATAACGTGGAAC$
testSample_1_2  141     *       0       0       *       *       0       0       AATGGAAGACCAGATGATTCTACTATATGACTCCAGACTAAGACAAATGCTAGGCTTTGAAAACGCA

[swbarnes2]

My guess is that the program is picky about the read names. You'd have to check the source itself to be sure.

Either the name of read one and read 2 should be the same, or read 1 should end in '/1', and read 2 in '/2'. Try those. The Picard suite also has a program to turn bams to fastqs.

[maubp]

As swbarnes2 points out, for a SAM/BAM file both parts of a pair of reads should be recorded with the same template name in column 1 (the suffix /1 or /2 or whatever can optionally be recorded in the tags). The FLAG in column 2 specifies which is the first read and which is the second.

Your filtered SAM file has four unique identifiers in column 1, therefore no complete pairs.

Double check the filter options you using with samtools view...

[Tomi]

Hi, thank you very much for your replies.

Yes, indeed the names were wrong, so I corrected it.
Basically I created a sample where I took the first 75bp of a reference chromosome, then I skipped a specific insert size and took the next 75, made the reverse and then the complementery strand out of it.

Fortunately I get the correct flags for the mapped reads, but I am still not able to export the reads where one is mapped and the second not - I mutated the second strand with a high mutation rate, just to make sure, that he can't map it.

Do you have an idea why? Is this not possible? I tried it with sam tools like that:
samtools view -b -S -f 8 -F 4 output.sam > mate.bam

and then with bam2fastq: bam2fastq -o mate#.fastq -f mate.bam

He is saying the correct number of mate reads (in that case one), but he still gets the warning I described above.


Here again the output of sam:

@SQ     SN:gi|157704448|ref|AC_000133.1|        LN:219475005
@PG     ID:bwa  PN:bwa  VN:0.5.9-r16
testSample_0    73      gi|157704448|ref|AC_000133.1|   1       37      75M     =       1       0       ATTGACAAGGGGAGGGAAAAGAGGAACAGAAATTCTTTTCTAT$
testSample_0    133     gi|157704448|ref|AC_000133.1|   1       0       *       =       1       0       TGGCTCTAACAGGCCACGATGGAATAGTCAATAATCACCTCTT$
testSample_1    99      gi|157704448|ref|AC_000133.1|   76      60      75M     =       226     225     AAATCCAGTTTGTGCCTACGGACATAATCTTTGAATTTGCTTT$
testSample_1    147     gi|157704448|ref|AC_000133.1|   226     60      75M     =       76      -225    AATAGATTTTCAAATAAGAAAATGAGAGGACATGAGCTTGAGG$
testSample_2    99      gi|157704448|ref|AC_000133.1|   301     60      75M     =       451     225     CTGACGACCTCCACGTGATTTCAACAATGATTTCAAATATTTC$
testSample_2    147     gi|157704448|ref|AC_000133.1|   451     60      75M     =       301     -225    TATAATCTATTGGCCATTCACAGCATAGCGTATAAACCTAGCT$

Thank you very much