search

AlexandrovLab / SigProfilerExtractorR

An R wrapper for SigProfilerExtractor that allows de novo extraction of mutational signatures from data generated in a matrix format. The tool identifies the number of operative mutational signatures, their activities in each sample, and the probability for each signature to cause a specific mutation type in a cancer sample. The tool makes use of SigProfilerMatrixGenerator and SigProfilerPlotting.
BSD 2-Clause "Simplified" License
15 stars 2 forks source link

trouble to extract mouse 'mm10' signatures through SigProfilerExtractorR #1

Closed luolingqi closed 3 years ago

luolingqi commented 4 years ago

Hi, I am new to the tool. I've tried to follow the 'easy to follow' instruction to install and run SigProfilerExtractorR on my VCF called against mouse (mm10). The simple R code that I used is:

sigprofilerextractor("vcf", "output", data, minsigs=1, maxsigs=3, replicates=10, cpu=-1, refgen = 'mm10',genome_build="mm10", mtype=c("DINUC,ID,96"))

But, the job_metadata file gives the following:

-------Vital Parameters Used for the execution ------- input_type: vcf inputdata: vcftest startProcess: 1 endProcess: 3 totalIterations: 10 cpu: -1 hierarchy: False refgen: GRCh38 genome_build: GRCh38 mtype: ['default'] init: random

At least 2 things are out of expectation here: 1) I specified 'mm10', which I successfully installed I believe, but here says 'GRCh38'; 2) I specified the mutation types that I want, but it still works on the 'default' mtype.

I don't have this issue when I run SigProfilerMatrixGeneratorR/PlottingR. Can anybody help? Plus, not sure if it is due to the above I always get an error message as below:

**Time taken to collect 10 iterations for 1 signatures is 0.55 seconds Optimization time is 0.01352691650390625 seconds The reconstruction error is 0.0445, average process stability is 1.0 and the minimum process stability is 1.0 for 1 signatures

Extracting signature 2 for mutation type 96 Error in py_call_impl(callable, dots$args, dots$keywords) : IndexError: index 1 is out of bounds for axis 1 with size 1

Detailed traceback: File "/usr/local/lib/python3.6/site-packages/sigproextractor/sigpro.py", line 571, in sigProfilerExtractor gpu=gpu,) File "/usr/local/lib/python3.6/site-packages/sigproextractor/subroutines.py", line 984, in decipher_signatures results = parallel_runs(genomes=genomes, totalProcesses=totalProcesses, iterations=totalIterations, n_cpu=cpu, verbose = False, resample=resample, seeds = seeds, init=init, normalization_cutoff=normalization_cutoff, gpu=gpu) File "/usr/local/lib/python3.6/site-packages/sigproextractor/subroutines.py", line 461, in parallel_runs result_list = pool.map(pool_nmf, seeds) File "/usr/local/lib/python3.6/multiprocessing/pool.py", line 266, in map return self._map_async(func, iterable, mapstar, chunksize).get() File "/usr/local/lib/python3.6/multiprocessing/pool.py", line 644, in get raise self._value**

Thanks!

luolingqi commented 4 years ago

It is not an issue for python version "SigProfilerExtractor". :-)

mishugeb commented 4 years ago

Hi, we are currently working on a new release for the R version. I will let you know soon after we release the version.