SEQanswers

Go Back   SEQanswers > Search Forums


Showing results 1 to 19 of 19
Search took 0.00 seconds.
Search: Posts Made By: sdriscoll
Forum: Bioinformatics 12-19-2014, 04:22 PM
Replies: 347
Views: 185,344
Posted By sdriscoll
Yeah they came up with some kind of so-called...

Yeah they came up with some kind of so-called perfect hash so that every kmer lookup is O(1). The slowest part of their program is probably the EM part (it slows down with increasing size of...
Forum: Bioinformatics 12-19-2014, 04:04 PM
Replies: 347
Views: 185,344
Posted By sdriscoll
Yeah that last point is totally fair. I'm also...

Yeah that last point is totally fair. I'm also not totally sure that for most applications allowing a mismatch really tells me what I want to know anyways.

On another subject, are you familiar...
Forum: Bioinformatics 12-19-2014, 03:21 PM
Replies: 347
Views: 185,344
Posted By sdriscoll
Hey Brian, bbduk and seal are certainly...

Hey Brian,

bbduk and seal are certainly great programs and since I have downloaded these tools I've been using them pretty much daily. While most kmer tools ignore the idea of allowing a mismatch...
Forum: Bioinformatics 12-16-2014, 02:35 PM
Replies: 347
Views: 185,344
Posted By sdriscoll
Yes, "only matching that sequence" is what I...

Yes, "only matching that sequence" is what I mean. So in seal if you use ambig=all then that read (hit) is added to all targets it matched (if there were multiple best matches)?

Or say...what...
Forum: Bioinformatics 12-16-2014, 01:11 PM
Replies: 347
Views: 185,344
Posted By sdriscoll
For seal, would there be any logical way to add...

For seal, would there be any logical way to add in number of unique hits to the rpkm= output? I'd also be interested to know what it does when a read has equally good matches to two or more target...
Forum: Bioinformatics 12-16-2014, 12:34 PM
Replies: 347
Views: 185,344
Posted By sdriscoll
OK, thanks that's useful. I guess logically if...

OK, thanks that's useful. I guess logically if I'm also specifying mink=12 and I only want reads that were trimmed I could also use maxlength=[read_length - mink] and then out would contain only...
Forum: Bioinformatics 12-16-2014, 11:37 AM
Replies: 347
Views: 185,344
Posted By sdriscoll
Yes. What option should I used so that I can get...

Yes. What option should I used so that I can get a FASTQ file with ONLY the reads that were trimmed?
Forum: Bioinformatics 12-16-2014, 10:54 AM
Replies: 347
Views: 185,344
Posted By sdriscoll
Hi Brian, I'm using bbduk to trim adapters...

Hi Brian,

I'm using bbduk to trim adapters from a microRNA seq run and in this case I only want to keep reads that were successfully trimmed. My command line is something like this:

bbduk.sh...
Forum: Bioinformatics 12-09-2014, 02:21 PM
Replies: 347
Views: 185,344
Posted By sdriscoll
Brian, Do you foresee making it possible to...

Brian,

Do you foresee making it possible to use kmers longer than 31bp at any point? Just curious.

Also could you add the 'trd' parameter to seal so that when one enables the 'rename' option...
Forum: Bioinformatics 11-24-2014, 04:22 PM
Replies: 347
Views: 185,344
Posted By sdriscoll
Oh that sounds fun.

Oh that sounds fun.
Forum: Bioinformatics 11-24-2014, 01:58 PM
Replies: 347
Views: 185,344
Posted By sdriscoll
Sorry, what is Seal?

Sorry, what is Seal?
Forum: Bioinformatics 11-21-2014, 11:01 AM
Replies: 347
Views: 185,344
Posted By sdriscoll
It seems like using mm=t may be enough for me in...

It seems like using mm=t may be enough for me in this case. I assume that means I'll be missing potential mismatches within K/2-1 of the ends of reads though. I guess I can live with that.
Forum: Bioinformatics 11-21-2014, 10:48 AM
Replies: 347
Views: 185,344
Posted By sdriscoll
I believe you can do it. Also I think you...

I believe you can do it.

Also I think you should call these things BBWeapons and not BBTools. They are much more like Weapons. Maybe one of my favorite tricks is this one:


samtools view -F...
Forum: Bioinformatics 11-21-2014, 10:38 AM
Replies: 347
Views: 185,344
Posted By sdriscoll
Oh yeah, sorry the -Xmx flag of course. I have a...

Oh yeah, sorry the -Xmx flag of course. I have a 3 year old and a 7 week old baby so I'm not all there all the time.

Something that may be useful is a feature like that in some other tools which...
Forum: Bioinformatics 11-21-2014, 09:54 AM
Replies: 347
Views: 185,344
Posted By sdriscoll
Hey Brian, When we use hdist=1 with BBDuk is...

Hey Brian,

When we use hdist=1 with BBDuk is the increase in memory use proportional to the reference or the input? If I keep getting "java.lang.OutOfMemoryError: Java heap space" errors then...
Forum: Bioinformatics 11-14-2014, 11:23 AM
Replies: 347
Views: 185,344
Posted By sdriscoll
These tools do so many things. The BBTools are...

These tools do so many things. The BBTools are my new favorite toys.
Forum: Bioinformatics 11-14-2014, 10:07 AM
Replies: 347
Views: 185,344
Posted By sdriscoll
This is great. Sorry there are so many tools in...

This is great. Sorry there are so many tools in the bbmap folder I haven't explored them all yet. Can't the set subtraction be handled by BBDuk? So if you run sequence set A and B separately through...
Forum: Bioinformatics 11-12-2014, 12:06 PM
Replies: 347
Views: 185,344
Posted By sdriscoll
Brian, Would it be possible for BBDuk to...

Brian,

Would it be possible for BBDuk to optionally export the matched kmers and unmatched (from input not reference) kmers? In a way it would be like generating a kmer library with something...
Forum: Bioinformatics 11-12-2014, 10:46 AM
Replies: 347
Views: 185,344
Posted By sdriscoll
Hi Brian (and anyone else in the know), Is...

Hi Brian (and anyone else in the know),

Is there a magic kmer length that is optimal for contaminant filtering? I know from short read alignment there are many fewer possible alignments for a 100b...
Showing results 1 to 19 of 19

 


All times are GMT -8. The time now is 02:49 AM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2021, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO