[GenoMax]

I am puzzled by why a miRNA dataset was sequenced as PE.

Can you check to see if those two reads overlap in the middle (you can use bbmerge.sh from BBMap suite: http://seqanswers.com/forums/showthread.php?t=43906)?

If this is starting to get confusing, just use one of the samples and its *_1.fastq.gz file for initial trial alignment.

[mastercoder]

I am puzzled by why a miRNA dataset was sequenced as PE.

Can you check to see if those two reads overlap in the middle (you can use bbmerge.sh from BBMap suite: http://seqanswers.com/forums/showthread.php?t=43906)?

If this is starting to get confusing, just use one of the samples and its *_1.fastq.gz file for initial trial alignment.

I have no idea why they did it as PE. Tomorrow first thing I am check BBMap and the documents you gave me. Again bunch of thanks. May your wishes come true! You ve been a huge help.

[Brian Bushnell]

Incidentally, for running BBMerge on miRNAs, I suggest adding the flag "mininsert=17". The default is 35 which is too big for microRNAs.

[nucacidhunter]

Do you know what was the fragment size for this library?

Target size is 20-30nt which translates to 145-160 bp fragments from the library. Usually there are some larger or smaller inserts as well from non-target transcripts due to size selection.

[mastercoder]

Hello again,

Here is the result I got from bbmerge, but i do not know how to interpret, if these overlapping in the middle or not. I feel illiterate. I also would like to ask you guys any documentation or any advice on how can I improve my self. Because there is nobody who is good at bioinfo where I am.

[Brian Bushnell]

That's pretty much as expected... on average, your miRNAs were 30 bp long, and virtually all (97%) of your read pairs overlapped.

[GenoMax]

Hello again,

Here is the result I got from bbmerge, but i do not know how to interpret, if these overlapping in the middle or not. I feel illiterate. I also would like to ask you guys any documentation or any advice on how can I improve my self. Because there is nobody who is good at bioinfo where I am.

Go ahead and merge the R1/R2 files and then use the resulting single read representation for doing the mapping. Since you are mapping miRNA don't allow any gaps in the alignments.

You can use bbmap.sh for this purpose. I think @Brian has recommendations for miRNA alignment in bbmap thread or the help document included in bbmap distribution.

Note: If you want to be extra cautious you could pass the trimmed R1/R2 files through bbduk.sh with "tbe tpo" option and then merge them. That should get rid of any stray bases from adapters that may have remained behind.

[mastercoder]

That's pretty much as expected... on average, your miRNAs were 30 bp long, and virtually all (97%) of your read pairs overlapped.

Okay, i will move forward to do mapping with BBMap. Thanks again. By the way this overlapping means, both R1/R2 have almost the same sequences? Thanks again.

[Brian Bushnell]

Okay, i will move forward to do mapping with BBMap. Thanks again. By the way this overlapping means, both R1/R2 have almost the same sequences? Thanks again.

Yes, that's correct (in the overlapping region). Except that R2 is the reverse-complement of R1.