pairs out of sync!!!

[Brian Bushnell]

As long as the names are in normal Illumina format, repair.sh will work.

[emolinari]

Hi everyone!!!
I have some issues analyzing old data from RNA-seq (2x75, HiSeq2000):

My reads, overall good Phred score, are probably out of sync.
Left reads:
Input : 4000000
Mapped : 3350445 (83.8% of input)
of these: 295039 ( 8.8%) have multiple alignments (2672 have >20)
Right reads:
Input : 4000000
Mapped : 3344059 (83.6% of input)
of these: 294834 ( 8.8%) have multiple alignments (2693 have >20)
83.7% overall read mapping rate.

Aligned pairs: 2801036
of these: 268208 ( 9.6%) have multiple alignments
2792534 (99.7%) are discordant alignments
0.2% concordant pair alignment rate.


I trimmed them with Trimmomatic to try to recover pairing (it always worked in the past!) but didn't do the job at all.


Left reads:
Input : 3182782
Mapped : 2700943 (84.9% of input)
of these: 240957 ( 8.9%) have multiple alignments (2680 have >20)
Right reads:
Input : 3182782
Mapped : 2699714 (84.8% of input)
of these: 240771 ( 8.9%) have multiple alignments (2687 have >20)
Unpaired reads:
Input : 107691
Mapped : 91246 (84.7% of input)
of these: 4641 ( 5.1%) have multiple alignments (47 have >20)
84.8% overall read mapping rate.

Aligned pairs: 2291108
of these: 220643 ( 9.6%) have multiple alignments
2284191 (99.7%) are discordant alignments
0.2% concordant pair alignment rate.


I have no idea how to fix this myself...this data was produced almost a year ago by our core facility and I am pretty sure they've got rid of the original (pre-filtered) data. Is there any code I can use or even a pairing program anyone recommends? I have read something about make pairs on Github...


Any help is highly appreciated!
thanks!!!

Manu