Does Darts works to the outlier(unpaired) analysis?

[sitongc]

Hi,
I try to run Darts in my 1 v 3 group, but it always give me an error when I run bayers_infer:

rpy2.rinterface.rruntimeerror: error in { : task 1 failed - "non-finite value supplied by optim"

Here is my code:

Darts_BHT bayes_infer --rmats-count JC.raw.input.A5SS.txt --od ./ --annot fromGTF.A5SS.txt -t A5SS -c 0.05 --nthread 8

Do you have any suggestions?

Best,
Sitong

[zj-zhang]

Hi @sitongc It looks like your run failed because Darts can't estimate a valid group variance when one of your groups has only 1 sample. You may want to try adding the "--replicate-model pooled" to merge the replicates with 3 samples, and run a pooled model.
Darts_BHT bayes_infer --rmats-count JC.raw.input.A5SS.txt --od ./ --annot fromGTF.A5SS.txt -t A5SS -c 0.05 --nthread 8 --replicate-model pooled

[sitongc]

Hi @zj-zhang, thank you for your reply.

I try to run my analysis by your code, but it is still give me the same error.

Darts_BHT bayes_infer --rmats-count JC.raw.input.A5SS.txt --od ./ --annot fromGTF.A5SS.txt -t A5SS -c 0.05 --nthread 8 --replicate-model pooled
2022-08-17 18:09:17,143 - Darts_BHT.bayes_infer - INFO -
Coverting rMATS count to Darts format
2022-08-17 18:09:17,205 - Darts_BHT.bayes_infer - INFO -
input count=./A5SS.input.txt
2022-08-17 18:09:17,206 - Darts_BHT.bayes_infer - INFO -
output dir=./
2022-08-17 18:09:17,216 - Darts_BHT.bayes_infer - INFO -
detected input file has no replicates; your "replicate_model" or "estim_gVar" options will be ignored
2022-08-17 18:09:17,217 - Darts_BHT.bayes_infer - INFO -
using no-replicate model
no. of threads using = 8
|======================================================================| 100%/home/ec2-user/miniconda3/envs/darts/lib/python2.7/site-packages/rpy2/rinterface/init.py:186: RRuntimeWarning: Error in { : task 1 failed - "non-finite value supplied by optim"

warnings.warn(x, RRuntimeWarning)
Traceback (most recent call last):
File "/home/ec2-user/miniconda3/envs/darts/bin/Darts_BHT", line 152, in
main()
File "/home/ec2-user/miniconda3/envs/darts/bin/Darts_BHT", line 48, in main
BHT_wrapper.parser( args )
File "/home/ec2-user/miniconda3/envs/darts/lib/python2.7/site-packages/Darts_BHT-0.1.0-py2.7.egg/Darts_BHT/BHT_wrapper.py", line 231, in parser
File "/home/ec2-user/miniconda3/envs/darts/lib/python2.7/site-packages/Darts_BHT-0.1.0-py2.7.egg/Darts_BHT/BHT_wrapper.py", line 54, in run_darts_BHT_norep
File "/home/ec2-user/miniconda3/envs/darts/lib/python2.7/site-packages/rpy2/robjects/functions.py", line 178, in call
return super(SignatureTranslatedFunction, self).call(*args, **kwargs)
File "/home/ec2-user/miniconda3/envs/darts/lib/python2.7/site-packages/rpy2/robjects/functions.py", line 106, in call
res = super(Function, self).call(*new_args, **new_kwargs)
rpy2.rinterface.RRuntimeError: Error in { : task 1 failed - "non-finite value supplied by optim"

[zj-zhang]

OK this could be tricky as the error msg is not very helpful. Could you let me know if you were able to run the test dataset (link below) without a problem?
https://github.com/Xinglab/DARTS/tree/master/Darts_BHT/test_data