Software packages for next gen sequence analysis

[zee]

Looking good. This site is very useful. Will anybody in this community be getting together at the ISMB Toronto 2008 meeting?

[jkbonfield]

Originally Posted by RudyS

anybody have scoops on possible software upgrade (GAP5?) for Staden package?

I just noticed this forum - seems like a little gem :-)

Gap5 development is in progress, although it's slow as I keep getting distracted with SRF and ZTR trace formats. Currently there's not official public release of Gap5 itself, just of the text terminal viewer using some of the same code (see below).

So far the basic underlying storage and searching methods exist plus a basic contig editor. It's definitely fast and very efficient (both memory and cpu) compared to Gap4, but Gap4 has one key advantage - it's finished (as much as anything ever is)!

I should probably work on getting a publically useable release ready sometime, at least for testing purposes. The text-mode version which shares the same file format but is just a simple curses-based viewer can be downloaded from https://sourceforge.net/project/show...kage_id=256957 although it's a tad out of date.

James

[sci_guy]

Updated table. Please inform me if I have missed something.

[sparks]

Originally Posted by sci_guy

Updated table. Please inform me if I have missed something.

Hi, I think novocraft should be in align/assemble.

[sci_guy]

Good spotting. Updated.

[bioinfosm]

Originally Posted by sci_guy

Updated table. Please inform me if I have missed something.

THere is the NextGENe from Softgenetics that looks pretty good with the condensation of reads

[sci_guy]

Thanks Bioinfosm. Now updated.

[mchaisso]

Originally Posted by zee

Looking good. This site is very useful. Will anybody in this community be getting together at the ISMB Toronto 2008 meeting?

I'll give a short presentation on EULER-SR at the short read SIG.

[ScottC]

[Edit: Just noticed James' post above in response to the GAP5 query!]

[nclement]

There's another program as well for mapping short reads called gnumap (http://dna.cs.byu.edu/gnumap/) made to increase the accuracy with duplicate matches. Open source, creates viewable output (with Affy's Integrated Genome Browser), and produces results very similar to novocraft's.

[lh3]

Another good program: ZOOM! Zillions of oligos mapped, which is online now at Bioinformatics. ZOOM resembles Eland a lot, but it further improves the spaced-seed method. I think SOLiD read mapper also uses quite a similar strategy of spaced seed, but it indexes genome, while ZOOM indexes reads like Eland.

Also, I believe ZOOM is carefully engineered. Outperforming eland with a similar algorithm is non-trivial, even given the advantages in the algorithm. I just wonder how ZOOM-C/I/P may perform. The authors did not give benchmark.

[apfejes]

I've seen the manuscript of their ZOOM publication, which looks impressive, and followed up with someone at the company that produces it - I was told that the software wasn't yet available, and might not be for some period of time.

Unfortunately, I get the impression ZOOM might be vapourware for the forseable future - though if anyone knows more than I do, please feel free to correct me.

[lh3]

I do not know more that you, apfejes. I just wonder why they do not release the program when the benchmark is ready. This might be the best time for "short" read aligners. In a year time (probably half-a-year time), the length of Illumina reads will come to >70bp which will effectively cripple a lot of current programs, including maq unfortunately.

[apfejes]

Thanks - all I know is that when I spoke to my contact at Bioinformatics Solutions, they were using the core of the software to generate benchmarks, but were still working on the application itself. A lot of my questions weren't answered, unfortunately.

Either way, I agree, once we pass 64bp reads (I've heard we're going to start doing 72bp test runs next week), we're going to leave the realm of short read aligners and need to start dealing with medium length (100-1000bp) read aligners, anyhow. (I'm not sure what's in that space, though: blast, blat, exonerate?)

As an aside, I'll save the term "long read alignments" for when Pacific Biosciences releases their SMRT (single molecule - real time) sequencing machine. 5-25k reads are about as long as I expect are going to be necessary for any application, though at that point, you're probably better off doing assembly than alignments.

[zee]

I will be looking out for some benchmark data for > 70bp Illumina runs to test future versions of novoalign/novopaired on. We had anticipated this would be coming and have some ideas on how to handle these.
If anybody could arrange some , public or under NDA, then we'd like that very much to get something out to the community.

Originally Posted by lh3

I do not know more that you, apfejes. I just wonder why they do not release the program when the benchmark is ready. This might be the best time for "short" read aligners. In a year time (probably half-a-year time), the length of Illumina reads will come to >70bp which will effectively cripple a lot of current programs, including maq unfortunately.

[apfejes]

Hi zee,

I haven't seen the runs yet - but if you message me in a week or two, I might be able to let you know how things ran here.

Actually, make that early September - I'm on holiday for the last 2 weeks of august. (-:

[spirit]

Hi, lh3 and apfejes. Thanks for you opinion on ZOOM . I am one of the developers of ZOOM. In fact, we will release the command-line version of ZOOM next week.

We finished the part of ZOOM dealing with Illumina/Solexa data in January of this year. Due to some personal reason, we came back to it until May. Then we found the release of ABI SOLiD data. We wanted to support color space data in ZOOM. After that, we focused on the GUI part, manual, website... We are trying to release an efficient, useful and easy-to-use tool. So, sorry for letting you wait. The good news is it will appear next week.

Welcome to try ZOOM out. We'd like to provide any help we could.

Originally Posted by lh3

I do not know more that you, apfejes. I just wonder why they do not release the program when the benchmark is ready. This might be the best time for "short" read aligners. In a year time (probably half-a-year time), the length of Illumina reads will come to >70bp which will effectively cripple a lot of current programs, including maq unfortunately.

[apfejes]

Thanks for the update, spirit.

Maybe you could give us a little bit of information on Zoom, as well, since things may have changed since last time I heard anything about it.

What are the longest and shortest reads it can handle effectively?
how does it compare to Eland or MAQ in reads aligned per minute?
How many mismatches does it handle?
Does it have a gapped mode?
What format is required for the reference genome?
What format are the alignments reported in?
Can you comment on the cost/licenses it will be provided under?
Can you give us the link to the download when it's ready?

I'm sure I'm missing other important information, but those are the first questions that occur to me.

Thanks!

[spirit]

And may I ask a question? What is the pair-end data of Illumina/Solexa data like? ZOOM now accepts two types of pair-end reads input. One is two fasta input files recording reads from two ends separately. ZOOM will automatically find the reads paired according to their name before mapping. The other is one fasta input files with two reads of a pair appearing in odd line and even line respectively. Thanks.

Originally Posted by lh3

Another good program: ZOOM! Zillions of oligos mapped, which is online now at Bioinformatics. ZOOM resembles Eland a lot, but it further improves the spaced-seed method. I think SOLiD read mapper also uses quite a similar strategy of spaced seed, but it indexes genome, while ZOOM indexes reads like Eland.

Also, I believe ZOOM is carefully engineered. Outperforming eland with a similar algorithm is non-trivial, even given the advantages in the algorithm. I just wonder how ZOOM-C/I/P may perform. The authors did not give benchmark.

[apfejes]

I suppose the PET comes out in the format provided by whichever base caller you use. I imagine it wouldn't be too difficult to convert to the formats ZOOM requires.

I believe that people here are starting to look at alternate base calling programs, so maybe someone who knows more than I do can point you in the direction of the documentation for those applications.