rMATS turbo v4.1.0 is available

[Eric Kutschera]

With this release the source code is now public: https://github.com/Xinglab/rmats-turbo

The full release notes are available at http://rnaseq-mats.sourceforge.net and the major changes from v4.0.2 are:

• Add command line arguments to run parts of the computation on different machines (--task, --tmp)
• Add command line argument to allow variable read lengths (--variable-read-length)
• Add command line argument for paired statistical analysis with PAIRADISE (--paired-stats)
• Add command line arguments to allow splicing events to be detected that involve an unannotated splice site (--novelSS, --mil, --mel). This is an experimental feature
• Compatible with both Python 2 and Python 3
• Only one sample group is required if --statoff is used
  

Please see the README for details about using the new version. Feel free to post any questions or feedback.

Thanks,

Eric

[Thomas Danhorn]

Thank you very much, I really appreaciate that! And perfect timimng, just
in time for a new project I am starting today.

Thomas

On Mon, 1 Jun 2020, Eric Kutschera wrote:

> With this release the source code is now public:
> https://github.com/Xinglab/rmats-turbo
>
> The full release notes are available at http://rnaseq-mats.sourceforge.net
> and the major changes from v4.0.2 are:
>

> - Add command line arguments to run parts of the computation on
> different machines (--task, --tmp)
> - Add command line argument to allow variable read lengths
> (--variable-read-length)
> - Add command line argument for paired statistical analysis with
> PAIRADISE (--paired-stats)
> - Add command line arguments to allow splicing events to be detected

> that involve an unannotated splice site (--novelSS, --mil, --mel). This is
> an experimental feature

> - Compatible with both Python 2 and Python 3
> - Only one sample group is required if --statoff is used
>
> Please see the README
> <https://github.com/Xinglab/rmats-turbo/blob/v4.1.0/README.md> for details

> about using the new version. Feel free to post any questions or feedback.
>
> Thanks,
> Eric
>

> --
> You received this message because you are subscribed to the Google Groups "rMATS User Group" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to rmats-user-gro...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/c201e6e5-ef0e-4dfd-a87b-e14b33047c57%40googlegroups.com.
>

[Hadas Ner-Gaon]

Those are great news. Thank you very much.

conda is still installing rmats 4.0.2

when it will be possible to conda install rmats 4.1.0?  

Hadas

בתאריך יום שני, 1 ביוני 2020 בשעה 23:21:13 UTC+3, מאת Thomas Danhorn:

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

[Thomas Danhorn]

The install scripts for 4.1.0 allow the use of coda. I don't use
coda and have not tried, but it looks like you can download the tar
ball from github, extract, and run `build_rmats --coda'. There is also a
test_rmats script that requires coda.

Hope this helps,

Thomas

>> an email to rmats-us...@googlegroups.com <javascript:>.

>>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/rmats-user-group/c201e6e5-ef0e-4dfd-a87b-e14b33047c57%40googlegroups.com.
>>
>>>
>>
>

> --
> You received this message because you are subscribed to the Google Groups "rMATS User Group" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to rmats-user-gro...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/be7061b9-91a0-4b89-b230-ad9f19f2a187%40googlegroups.com.
>

[Eric Kutschera]

As Thomas mentioned, the github release has a script that creates a conda environment and installs rmats to that environment: `./build_rmats --conda`.

We are still working on getting a bioconda package for 4.1.0 uploaded so that `conda install -c bioconda rmats` will install 4.1.0.

Eric

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

[Jeremy Leipzig]

Hi Eric,

You should probably clarify if reads of variable length are actually weighed as such in the statistics or you simply allowed users to turn an alarm off.

"Allow reads with lengths that differ from --readLength to be processed. --readLength will still be used to determine IncFormLen and SkipFormLen."

This really means (I assume) the program will just pretend reads shorter or longer than readLength will be treated as if they are readLength, which is kind of dangerous if people don't grasp that import of that statement.

So maybe just adding "Use this flag at your own risk!" would alert people that they may be about to shoot themselves in the foot.

-j

--
You received this message because you are subscribed to the Google Groups "rMATS User Group" group.

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

[YI XING]

Hi Jeremy,

Thanks for your comment. We actually did extensive assessments of this issue using simulated and real datasets, and found that the effect of the read length parameter on splicing quantitation is extremely minor. Even when we take a dataset with uniform read length, and intentionally set a wrong read length in the input parameter, the splicing quantitation almost doesn't change at all. We think this is because the effect of the read length parameter gets canceled out in both numerator and denominator during the splicing quantitation. With that, we feel pretty comfortable to provide this option to users.

Yours,

Yi

To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/CAAiprghuNfLDJPiqOeZ67noB6NmEEiMfhnXobD1ESGJ1vag-kg%40mail.gmail.com.

[Eric Kutschera]

4.1.0 is now available on bioconda: https://anaconda.org/bioconda/rmats

Eric

[严锋]

Thanks so much for the release.

But using the latest conda (4.8.3), I still get 4.0.2 by `conda install -c bioconda rmats`.

If I change it to rmats=4.1.0, it says conflicts found and looking for incompatible packages.

Have you had a successful example or code?

Thank you 

Alex

在 2020年6月5日星期五 UTC+10下午11:47:09，Eric Kutschera写道：

[Eric Kutschera]

The 4.1.0 package on bioconda seems to require the conda-forge channel. You should be able to install 4.1.0 with this command:

  conda install -c conda-forge -c bioconda rmats

That tells conda to also use conda-forge for dependencies. Here's the documentation from Bioconda about requiring conda-forge: https://bioconda.github.io/user/install.html#set-up-channels

Eric

[Feng Alex Yan]

Thanks, it might be some dependencies required the conda-forge channel.

在 2020年6月9日星期二 UTC+10下午11:04:07，Eric Kutschera写道：

[Lorebg]

Hi, 

I'm installing rmats turbo v4.1.0 with conda using:

conda install -c conda-forge -c bioconda rmats

everything looks fine in my environment and I can see rmats 4.1.0 has been installed

I then run rmats with:

 

python path_to_my_env/rMATS/rmats.py {arguments}

but I get the following error:

Traceback (most recent call last):

  File "/path_to_my_env/rMATS/rmats.py", line 431, in <module>

    main()

  File "/path_to_my_env/rMATS/rmats.py", line 396, in main

    args = get_args()

  File "/path_to_my_env/rMATS/rmats.py", line 158, in get_args

    args.b1 = fp.read().strip(' ,\n')

  File "path_to_my_env/lib/python3.7/codecs.py", line 322, in decode

    (result, consumed) = self._buffer_decode(data, self.errors, final)

UnicodeDecodeError: 'utf-8' codec can't decode byte 0x8b in position 1: invalid start byte

Can someone help with this error and how can I fix it? Thanks

Lorea

[Thomas Danhorn]

If I am interpreting the error message correctly, the file you are passing
with the --b1 option contains an invalid character. In case your editor
introduces garbage, you could recreate that file with something like:

echo '/path/to/bam1.bam,/path/to/bam2.bam,/path/to/bam3.bam' > b1.txt

Good luck!

Thomas

> To unsubscribe from this group and stop receiving emails from it, send an email to rmats-user-gro...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/b7b87c59-2289-40d7-b69d-4a0775c2cabbo%40googlegroups.com.
>

[Lorebg]

Yes, it worked! My bad, it was my first time running the turbo version and I set the path to .bam files as in v3.2.5

Once thing I noticed is that there is not summary.txt file as in v3.2.5, which was quite useful to have a general idea of the analysis. Is this something which can be added somehow?

Thanks, 

Lorea

> To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/b7b87c59-2289-40d7-b69d-4a0775c2cabbo%40googlegroups.com.
>

[Eric Kutschera]

The information that was in the summary.txt file from v3.2.5 is easy to generate from the other output files in v4.1.0. You can write a small script that reads the SE.MATS.JC.txt and similar output files and that counts the events which meet your desired criteria for the PValue, FDR, and IncLevel* columns.

Eric

[Eric Kutschera]

I'll work on adding a script that generates a summary.txt file to https://github.com/Xinglab/rmats-turbo

Eric

[Lorebg]

Thanks, that'd be really helpful :)

Lorea

[Eric Kutschera]

I created a script to generate a summary.txt. Here is the pull request: https://github.com/Xinglab/rmats-turbo/pull/10. Please take a look and post any feedback as comments on the pull request

Eric

[Thomas Danhorn]

I suggest renaming the --cutoff option to --p-cutoff (or --fdr-cutoff or
similar), in case you want to implement (either now or later) an
IncLevelDiff cutoff.

I also think most people would (and should) be more interested in FDR than
raw p-values, so maybe make the --fdr behavior the default and have a
--raw-p (or similar) option.

> To unsubscribe from this group and stop receiving emails from it, send an email to rmats-user-gro...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/b21c4f57-479a-4b74-89d1-f929a343c145o%40googlegroups.com.
>

[Eric Kutschera]

Thanks, those are good suggestions. I've renamed to --p-cutoff, implemented an IncLevelDiff cutoff, and now check FDR by default. The pull request is updated

Eric

> To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/b21c4f57-479a-4b74-89d1-f929a343c145o%40googlegroups.com.
>

[Thomas Danhorn]

Looks great, thanks! One more thought just popped into my head -- since
we can set different cutoffs, the user might want to generate more than
one summary, e.g. one with really strict cutoffs, and one with more
lenient ones. For this it might be convenient to either have the option
to rename the output file, or maybe to incorporate the
FDR/p-value/IncLevelDiff cutoffs into the file name (e.g.
Summary_FDR0.05_ILD0.txt, or Summary_p0.001_ILD0.1.txt).

Thomas

>> https://groups.google.com/d/msgid/rmats-user-group/b21c4f57-479a-4b74-89d1-f929a343c145o%40googlegroups.com.

>>
>>>
>
> --
> You received this message because you are subscribed to the Google Groups "rMATS User Group" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to rmats-user-gro...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/90608823-5078-4090-a9d9-3b37404808d6o%40googlegroups.com.
>

[Eric Kutschera]

Providing options for naming the summary file is also a good suggestion. I've updated the pull request

Eric

> To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/90608823-5078-4090-a9d9-3b37404808d6o%40googlegroups.com.
>

[LOREA BLAZQUEZ]

Hi Eric, 

Thank you very much for writing the summary.py script. I couldn't have a look until today, but it works perfectly fine

Best, 

Lorea

To unsubscribe from this group and stop receiving emails from it, send an email to rmats-user-gro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/3fc6c93c-7cdb-4080-b270-f3c5cb2aa8f1o%40googlegroups.com.

[Matthew Romero]

Hello, I am new to rMATS and am trying out rMATS-turbo using Ubuntu. I was able to install the necessary dependencies and build the rmats and test_rmats environments; however, after I have activated the test environment and run the automated tests with "./run_rmats" I keep getting the error:

CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'.

To initialize your shell, run

    $ conda init <SHELL_NAME>

Currently supported shells are:

  - bash

  - fish

  - tcsh

  - xonsh

  - zsh

  - powershell

See 'conda init --help' for more information and options.

IMPORTANT: You may need to close and restart your shell after running 'conda init'.

I have tried the suggested solutions above and searched for others but nothing seems to work. I was wondering if anyone could tell me the solution to this error? Does it have to do with how rMATS was installed/built? Or am I missing some sort of package for general usage? 

Any help would be much appreciated.

[Thomas Danhorn]

If you did not use the conda installer, you don't need run_rmats (it just
sets up the condo environment before calling rmats.py) -- use rmats.py
directly as the executable instead. (If you want to call it as "rmats.py
<parameters>", you will probably need to make it executable with "chmod
+x /path/to/rmats.py"; but you can always call it as "python
/path/to/rmats.py <parameters>").

Hope this helps,

Thomas

On Mon, 6 Jul 2020, Matthew Romero wrote:

> Hello, I am new to rMATS and am trying out rMATS-turbo using Ubuntu. I was
> able to install the necessary dependencies and build the rmats and
> test_rmats environments; however, after I have activated the test
> environment and run the automated tests with "./run_rmats" I keep getting

> the *error:*

>
> CommandNotFoundError: Your shell has not been properly configured to use
> 'conda activate'.
> To initialize your shell, run
>
> $ conda init <SHELL_NAME>
>
> Currently supported shells are:
> - bash
> - fish
> - tcsh
> - xonsh
> - zsh
> - powershell
>
> See 'conda init --help' for more information and options.
>
> IMPORTANT: You may need to close and restart your shell after running
> 'conda init'.
>
> I have tried the suggested solutions above and searched for others but
> nothing seems to work. I was wondering if anyone could tell me the solution
> to this error? Does it have to do with how rMATS was installed/built? Or am
> I missing some sort of package for general usage?
> Any help would be much appreciated.
>
> On Monday, June 1, 2020 at 12:30:49 PM UTC-7, Eric Kutschera wrote:
>>
>> With this release the source code is now public:
>> https://github.com/Xinglab/rmats-turbo
>>
>> The full release notes are available at http://rnaseq-mats.sourceforge.net
>> and the major changes from v4.0.2 are:
>>

>> - Add command line arguments to run parts of the computation on
>> different machines (--task, --tmp)
>> - Add command line argument to allow variable read lengths
>> (--variable-read-length)
>> - Add command line argument for paired statistical analysis with
>> PAIRADISE (--paired-stats)
>> - Add command line arguments to allow splicing events to be detected

>> that involve an unannotated splice site (--novelSS, --mil, --mel). This is
>> an experimental feature

>> - Compatible with both Python 2 and Python 3
>> - Only one sample group is required if --statoff is used
>>
>> Please see the README
>> <https://github.com/Xinglab/rmats-turbo/blob/v4.1.0/README.md> for

>> details about using the new version. Feel free to post any questions or
>> feedback.
>>
>> Thanks,
>> Eric
>>
>>
>

> --
> You received this message because you are subscribed to the Google Groups "rMATS User Group" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to rmats-user-gro...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/5c71950e-b700-4b42-b580-abbf57772fcao%40googlegroups.com.
>

[Matthew Romero]

Hello again, thank you for your quick reply to my first problem.

I have been setting up rmats-turbo on a HPC computing cluster, and am encountering a new error: when running /path/to/env/rmats.py {arguments} , I get the error

"  File "rmats.py", line 16, in <module>

    from rmatspipeline import run_pipe

ModuleNotFoundError: No module named 'rmatspipeline' "

I downloaded all the dependencies and successfully ran ./build rmats --conda, in my installation and was able to activate the rmats environment, I was wondering how I might fix this? 

On Monday, June 1, 2020 at 12:30:49 PM UTC-7, Eric Kutschera wrote:

With this release the source code is now public: https://github.com/Xinglab/rmats-turbo

The full release notes are available at http://rnaseq-mats.sourceforge.net and the major changes from v4.0.2 are:

• Add command line arguments to run parts of the computation on different machines (--task, --tmp)

• Add command line argument to allow variable read lengths (--variable-read-length)

• Add command line argument for paired statistical analysis with PAIRADISE (--paired-stats)

• Add command line arguments to allow splicing events to be detected that involve an unannotated splice site (--novelSS, --mil, --mel). This is an experimental feature

• Compatible with both Python 2 and Python 3

• Only one sample group is required if --statoff is used
  

Please see the README for details about using the new version. Feel free to post any questions or feedback.

Thanks,

Eric

[Thomas Danhorn]

Do you have the environment set up correctly? E.g., the variable
PYTHONPATH has to point to the relevant site-packages directory where you
installed all Python dependencies.

On Mon, 13 Jul 2020, Matthew Romero wrote:

> Hello again, thank you for your quick reply to my first problem.
> I have been setting up rmats-turbo on a HPC computing cluster, and am
> encountering a new error: when running /path/to/env/rmats.py {arguments} ,
> I get the error
> " File "rmats.py", line 16, in <module>
> from rmatspipeline import run_pipe
> ModuleNotFoundError: No module named 'rmatspipeline' "
>
> I downloaded all the dependencies and successfully ran ./build rmats
> --conda, in my installation and was able to activate the rmats environment,
> I was wondering how I might fix this?
>
> On Monday, June 1, 2020 at 12:30:49 PM UTC-7, Eric Kutschera wrote:
>>
>> With this release the source code is now public:
>> https://github.com/Xinglab/rmats-turbo
>>
>> The full release notes are available at http://rnaseq-mats.sourceforge.net
>> and the major changes from v4.0.2 are:
>>

>> - Add command line arguments to run parts of the computation on
>> different machines (--task, --tmp)
>> - Add command line argument to allow variable read lengths
>> (--variable-read-length)
>> - Add command line argument for paired statistical analysis with
>> PAIRADISE (--paired-stats)
>> - Add command line arguments to allow splicing events to be detected

>> that involve an unannotated splice site (--novelSS, --mil, --mel). This is
>> an experimental feature

>> - Compatible with both Python 2 and Python 3
>> - Only one sample group is required if --statoff is used
>>
>> Please see the README
>> <https://github.com/Xinglab/rmats-turbo/blob/v4.1.0/README.md> for

>> details about using the new version. Feel free to post any questions or
>> feedback.
>>
>> Thanks,
>> Eric
>>
>>
>

> --
> You received this message because you are subscribed to the Google Groups "rMATS User Group" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to rmats-user-gro...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/914e562f-c3b8-4fcd-a511-90e945820903o%40googlegroups.com.
>

[Matthew Romero]

I followed all the instructions posted when setting up the environment, but I am not sure. How would I check my PYTHONPATH and update it to the correct directory?

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

[Eric Kutschera]

I think the issue is that you are trying to run with "/path/to/env/rmats.py" which ends up using a different version of python than the one created by "./build_rmats --conda". Do you get the same error with "./run_rmats"? You could also try activating the conda enviroment and then running "python /path/to/env/rmats.py"

Eric

[Matthew Romero]

I tried both methods and get the same error in either case.

[Thomas Danhorn]

To check the PYTHONPATH, see what

echo "$PYTHONPATH"

shows you. But if everything you installed is in the default Python
directory, you won't need to set PYTHONPATH, and rmats.py should be to
find its own rMATS installion ...


find /your/rMATS.4.1.0_installion_directory rmatspipeline

would show you, if the package rmatspipeline is actually installed where
it should be.

>> an email to rmats-us...@googlegroups.com <javascript:>.

>>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/rmats-user-group/914e562f-c3b8-4fcd-a511-90e945820903o%40googlegroups.com.
>>
>>>
>>
>

> --
> You received this message because you are subscribed to the Google Groups "rMATS User Group" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to rmats-user-gro...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/rmats-user-group/da558b87-24ea-4fd5-bc8e-b4116ee8fda5o%40googlegroups.com.
>

[Matthew Romero]

The only thing in PYTHONPATH was /opt/sdsc/lib.

Strange enough, when I run "rmats.py" by itself without activating the rmats environment while in the correct directory it seems to work. Is this because I installed in the base environment?

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

[Matthew Romero]

Hello, I am planning on running rmats on a dataset within an HPC computing cluster using the paired model for two illumina datasets (readlength 150bp, paired end reads, 8GB gzipped each). And I was wondering what the optimal number of threads and the required walltime for a such a job would be? Or what some resources are that might help me figure this out? 

Thanks again.

On Monday, June 1, 2020 at 12:30:49 PM UTC-7, Eric Kutschera wrote:

With this release the source code is now public: https://github.com/Xinglab/rmats-turbo

The full release notes are available at http://rnaseq-mats.sourceforge.net and the major changes from v4.0.2 are:

• Add command line arguments to run parts of the computation on different machines (--task, --tmp)

• Add command line argument to allow variable read lengths (--variable-read-length)

• Add command line argument for paired statistical analysis with PAIRADISE (--paired-stats)

• Add command line arguments to allow splicing events to be detected that involve an unannotated splice site (--novelSS, --mil, --mel). This is an experimental feature

• Compatible with both Python 2 and Python 3

• Only one sample group is required if --statoff is used
  

Please see the README for details about using the new version. Feel free to post any questions or feedback.

Thanks,

Eric

[Eric Kutschera]

rMATS can process each BAM file on a separate thread so you could use as many threads as you have BAM files. For large data sets running the prep and post steps separately can help: https://github.com/Xinglab/rmats-turbo#running-prep-and-post-separately

I'm not sure what the walltime would be. I would just try running it to see how long it takes. If you want to be able to predict the run time you could try running different sized subsets of your data and seeing how the walltime changes with the input size

Eric