Announcement: Trinity Release v2.10.0

[Brian Haas]

Greetings all,

We have a new Trinity release available.

https://github.com/trinityrnaseq/trinityrnaseq/releases/tag/v2.10.0

Those that are using the Trinity genome-guided pipeline should notice runtime and accuracy performance improvements.

Docker and singularity images are updated.

Full changelog notes:

# Trinity-v2.10.0 Mar 18, 2020

-added bamsifter for genome-guided Trinity-based aligned read normalization pre-assembly

-DTU updates for py3

-docker/singularity updates: now uses R-3.6.3 and py3

-improved error handling and test coverage in trinity-seqtk

-can specify read groups in variant detection pipeline runner

-kmer size can be adjusted again for experimental purposes only

-R less verbose on exec

-minor bugfixes

[gutie300 University of Minnesota]

Hi,

I am not able to change the kmer size, as I was in other previous versions.

For instance:

--KMER_SIZE 31

Gives the error:

ERROR, don't recognize parameter: --KMER_SIZE
Please review usage info for accepted parameters.

I thought that in this new version kmer size could be modified.

Thanks

[Brian Haas]

Yes, that's right. It's not recommended that you change it.  If you must for whatever reason, you can use --__KMER_SIZE

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/trinityrnaseq-users/b378a5c0-543d-442c-9181-e2f6feb3eec6%40googlegroups.com.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas

 

[repick chen]

Hi, brian

I am using Trinity to assemble transcriptomes from 21 samples. Since I cannot assemble the reads from all sample into a single Trinity run which consumes LOTS of RAM, I am assembling each data set independently. 

I want to merge all assemblies (21 trinity.fasta files) into a single, unified assembly. I have seen that there are several tools that aim to do this ( Corset, TransFuse, transrate).  I do not know which one would be better.

Any suggestions.

Thanks a lot.

Repick.

[repick chen]

[Brian Haas]

Hi Repick,

I think you should be able to assemble them all into a single assembly, assuming the reads all come from the same organism.  The memory usage has to do with the complexity of the sample, and if the reads are all derived from the same target organism, then you shouldn't reach an impassable level of complexity (unless there's contamination). If you use trimmomatic and ensure that the reads get normalized, I expect it should be fine.  I'm assuming you have access to a machine that has lots of RAM, though.  If you are RAM limited (ie. < 256G) then yes, you could run into trouble.

If you do need to run them separately and combine them, I'm not sure what's best.   This was published recently:
https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-020-6528-x

and should mention the various challenges and alternatives.  I haven't tried it yet myself...  (too many other projects I'm focused on atm).

best,

~b

--

You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/trinityrnaseq-users/fbd07500-b29d-4a25-afaa-4f1158de5a18%40googlegroups.com.

[repick chen]

Hi, Brian

Thanks so much for your suggestion.

It is troublesome that my RAM is 128g.  Is there some paratmeter setting could be used to minimize the use of RAM. 

Before i assemble  the 21 samples seperately, i have to try to assemble all the sample into a single assemly. It is pity that the process stuck at "-reading Kmer occurrences..."  

And i have used  trimmomatic and the normalization is on by default.

Is there any other ways i can do?

Thanks again.

Repick

在 2020年4月9日星期四 UTC+8下午9:56:46，Brian Haas写道：

Hi Repick,

I think you should be able to assemble them all into a single assembly, assuming the reads all come from the same organism.  The memory usage has to do with the complexity of the sample, and if the reads are all derived from the same target organism, then you shouldn't reach an impassable level of complexity (unless there's contamination). If you use trimmomatic and ensure that the reads get normalized, I expect it should be fine.  I'm assuming you have access to a machine that has lots of RAM, though.  If you are RAM limited (ie. < 256G) then yes, you could run into trouble.

If you do need to run them separately and combine them, I'm not sure what's best.   This was published recently:
https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-020-6528-x

and should mention the various challenges and alternatives.  I haven't tried it yet myself...  (too many other projects I'm focused on atm).

best,

~b

On Thu, Apr 9, 2020 at 6:36 AM repick chen <zhongzai...@gmail.com> wrote:

Hi, brian

I am using Trinity to assemble transcriptomes from 21 samples. Since I cannot assemble the reads from all sample into a single Trinity run which consumes LOTS of RAM, I am assembling each data set independently. 

I want to merge all assemblies (21 trinity.fasta files) into a single, unified assembly. I have seen that there are several tools that aim to do this ( Corset, TransFuse, transrate).  I do not know which one would be better.

Any suggestions.

Thanks a lot.

Repick.

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsub...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/trinityrnaseq-users/fbd07500-b29d-4a25-afaa-4f1158de5a18%40googlegroups.com.

[Brian Haas]

To lower memory requirements, you can try including
     --min_kmer_cov 2

and that will help during the inchworm stage.    Usually, the earlier jellyfish kmer counting phase doesn't lock up.   The 128G RAM is definitely suboptimal, though.  Ideally, you'd have 256 to 512G RAM for large jobs.

If you need to run separate assemblies and then combine them, be sure to update the transcript accession names so ensure they're unique before doing the merge. Otherwise, you'll find many of the same accessions being used in the different assembly files.

best,

~b

To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/trinityrnaseq-users/fbd07500-b29d-4a25-afaa-4f1158de5a18%40googlegroups.com.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas

 

--

You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/trinityrnaseq-users/9f556d7d-3998-482e-a39b-3c24961c6776%40googlegroups.com.