alexandrovlab / sigprofilersimulator Goto Github PK

Hi @ebergstr,

Thanks for the great tool! A couple of suggestions from me:

1. The argument c("96") of contexts variable in the example sigSim.SigProfilerSimulator command in Readme throws an error in my environment: NameError: name 'c' is not defined
contexts=["96"] works though, so I suggest changing example command to that.

2. If the reference genome is not downloaded, nothing seems to happen when Chromosome proportion file does not exist. Creating now... message appears. Manual download as per SigProfilerMatrixGenerator Readme works, so perhaps it's worth mentioning it in SigProfilerSimulator readme, too.

3. When installing the reference genome, The ensembl ftp site is not currently responding. error appears even when WGET is not installed. Although this tool is mentioned in Readme as a prerequisite, perhaps it's worth adding an exception here to check if WGET is available.

Best,
Sergey

https://github.com/AlexandrovLab/SigProfilerSimulator/blob/ced39a81e38098fc9c42b90255e5f70144ac52aa/SigProfilerSimulator/SigProfilerSimulator.py#L162C1-L163C67

Because the chromosomes are hardcoded without the chr prefix, it is impossible to run SigProfilerSimulator on non-Ensembl GRCh38 genomes that have the "chr" prefix before the chromosome number. Additionally, there is no MT chromosome listed so the code also breaks on vcf files containing MT chromosomes.

Hello,

I am running your tool using:

sigSim.SigProfilerSimulator(name, \
  vcf_dir, \
  "GRCh37", \
  contexts=["96", "ID"], \
  exome=None, \
  simulations=1000, \
  updating=False, \
  bed_file=bed, \
  overlap=False, \
  gender='female', \
  chrom_based=True, \
  seed_file=None, \
  noisePoisson=False, \
  cushion=100, \
  region=None, \
  vcf=True)

But I am getting the following error:

multiprocessing.pool.RemoteTraceback: 
"""
Traceback (most recent call last):
  File "/gs/gsfs0/users/rcutler/.conda/envs/SigProfilerSimulator/lib/python3.12/multiprocessing/pool.py", line 125, in worker
    result = (True, func(*args, **kwds))
                    ^^^^^^^^^^^^^^^^^^^
  File "/gs/gsfs0/users/rcutler/.conda/envs/SigProfilerSimulator/lib/python3.12/site-packages/SigProfilerSimulator/mutational_simulator.py", line 973, in simulator
    random_sample = random.sample(list(mutation_tracker[context]),1)[0]
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/gs/gsfs0/users/rcutler/.conda/envs/SigProfilerSimulator/lib/python3.12/random.py", line 430, in sample
    raise ValueError("Sample larger than population or is negative")
ValueError: Sample larger than population or is negative
"""

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/gs/gsfs0/shared-lab/vijg-lab/2023-Ronnie/231009_multiple_ENU_analysis/SigProfilerSimulator/merged/runSigProfilerSimulator.py", line 10, in <module>
    sigSim.SigProfilerSimulator(name, \
  File "/gs/gsfs0/users/rcutler/.conda/envs/SigProfilerSimulator/lib/python3.12/site-packages/SigProfilerSimulator/SigProfilerSimulator.py", line 479, in SigProfilerSimulator
    r.get()
  File "/gs/gsfs0/users/rcutler/.conda/envs/SigProfilerSimulator/lib/python3.12/multiprocessing/pool.py", line 774, in get
    raise self._value
ValueError: Sample larger than population or is negative

When I run your tool with chrom_based=False I am able to get results. So this makes me think it is an error when wanting to have mutations simulated by chromosome. Since some of my samples don't have many mutations, I think this may be due to some chromosomes having 0 mutations. Any help with this?

Thanks,
Ronnie

Dear author.

Thanks for making this wonderful tool.
I have questions regarding this tool.

I would like to simulate mutations within genes.
In order to do this, I inputted gene regions in "bed_file".
(sigSim.SigProfilerSimulator("SIMUL", "./up_gene", "GRCh38", contexts=["96"],simulations=120,bed_file="./updown_bed_canonincal_hg38")
However, this tool seemed to simulate a subset of mutations with this parameter.
(e.g., the number of total mutations within genes is around 10,000, but, each maf file contains about 700 mutations.)

Next, I inputted "BED_GRCh38_proportions", which seemed to be generated when using "bed_file" parameter, in the region parameter. (sigSim.SigProfilerSimulator("SIMUL", "./up_gene", "GRCh38", contexts=["96"],simulations=120,region="BED_GRCh38_proportions")

However, this approach caused the following error.
Could you tell me the best way to simulate mutations within genes?

multiprocessing.pool.RemoteTraceback:
"""
Traceback (most recent call last):
File "/lustre/scratch117/casm/team78/hj6/anaconda3/lib/python3.7/multiprocessing/pool.py", line 121, in worker
result = (True, func(*args, **kwds))
File "/lustre/scratch117/casm/team78/hj6/anaconda3/lib/python3.7/site-packages/SigProfilerSimulator/mutational_simulator.py", line 2349, in simulator
mutNuc = ''.join([tsb_ref[base][1] for base in sequence[random_number - mut_start:random_number + mut_start+1]])
File "/lustre/scratch117/casm/team78/hj6/anaconda3/lib/python3.7/site-packages/SigProfilerSimulator/mutational_simulator.py", line 2349, in
mutNuc = ''.join([tsb_ref[base][1] for base in sequence[random_number - mut_start:random_number + mut_start+1]])
KeyError: 51
"""

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "", line 1, in
File "/lustre/scratch117/casm/team78/hj6/anaconda3/lib/python3.7/site-packages/SigProfilerSimulator/SigProfilerSimulator.py", line 479, in SigProfilerSimulator
r.get()
File "/lustre/scratch117/casm/team78/hj6/anaconda3/lib/python3.7/multiprocessing/pool.py", line 657, in get
raise self._value
KeyError: 51

Hello,

Thanks for a great tool! I have been using it to randomize the position of mutations in samples while controlling for mutational signatures. However, I have noticed that I am not getting the same amount of mutations in the simulated files as in the input, which I would expect.

I am running the tool like this, including some specific regions to simulate in (based on coverage):
`#!/usr/bin/env python

import sys
from SigProfilerSimulator import SigProfilerSimulator as sigSim

name = sys.argv[1]
vcf_dir = sys.argv[2]
bed = sys.argv[3].strip()

sigSim.SigProfilerSimulator(name,
vcf_dir,
"GRCh37",
contexts=["96", "ID"],
exome=None,
simulations=10,
updating=False,
bed_file=bed,
overlap=False,
gender='female',
chrom_based=False,
seed_file=None,
noisePoisson=False,
cushion=100,
region=None,
vcf=True)`

Here is the output of the log file:
`======================================
SigProfilerSimulator
Checking for all reference files and relevant matrices...
Creating a chromosome proportion file for the given BED file ranges...Completed!
SigProfilerSimulator_cell_output/C1_05_24/output/ID/C1_05_24.ID83.region does not exist. Creating the matrix file now.
Starting matrix generation for SNVs and DINUCs...Completed! Elapsed time: 116.12 seconds.
Starting matrix generation for INDELs...Completed! Elapsed time: 106.96 seconds.
Matrices generated for 1 samples with 0 errors. Total of 597 SNVs, 1 DINUCs, and 15 INDELs were successfully analyzed.
The context distribution file does not exist. This file needs to be created before simulating. This may take several hours...
Chromosome X done
Chromosome 1 done
Chromosome 2 done
Chromosome 3 done
Chromosome 4 done
Chromosome 5 done
Chromosome 6 done
Chromosome 7 done
Chromosome 8 done
Chromosome 9 done
Chromosome 10 done
Chromosome 11 done
Chromosome 12 done
Chromosome 13 done
Chromosome 14 done
Chromosome 15 done
Chromosome 16 done
Chromosome 17 done
Chromosome 18 done
Chromosome 19 done
Chromosome 20 done
Chromosome 21 done
The context distribution file has been created!
The context distribution file does not exist. This file needs to be created before simulating. This may take several hours...
Chromosome X done
Chromosome 1 done
Chromosome 2 done
Chromosome 3 done
Chromosome 4 done
Chromosome 5 done
Chromosome 6 done
Chromosome 7 done
Chromosome 8 done
Chromosome 9 done
Chromosome 10 done
Chromosome 11 done
Chromosome 12 done
Chromosome 13 done
Chromosome 14 done
Chromosome 15 done
Chromosome 16 done
Chromosome 17 done
Chromosome 18 done
Chromosome 19 done
Chromosome 20 done
Chromosome 21 done
The context distribution file has been created!

Files successfully read and mutations collected. Mutation assignment starting now.
Mutations have been distributed. Starting simulation now...
Chromosome 21 done
Chromosome 22 done
Chromosome 19 done
Chromosome 20 done
Chromosome 18 done
Chromosome 13 done
Chromosome 17 done
Chromosome 15 done
Chromosome 16 done
Chromosome 14 done
Chromosome 9 done
Chromosome 11 done
Chromosome 10 done
Chromosome 12 done
Chromosome X done
Chromosome 8 done
Chromosome 7 done
Chromosome 6 done
Chromosome 4 done
Chromosome 5 done
Chromosome 3 done
Chromosome 1 done
Chromosome 2 done
Simulation completed
Job took 2381.3919444084167 seconds`

In my input file I have 613 mutations, but in the simulated output I only get 313 mutations. This issue is not specific to this file. Any help with this? I have tried to circumvent this by creating many more simulations than I need and sampling them to get the matching number of mutations as in my real sample.

Thanks,
Ronnie

Description:
Recent changes in SigProfilerMatrixGenerator v1.3.18 have implications for SigProfilerSimulator:

• File SigProfilerMatrixGenerator.py has been renamed to MutationMatrixGenerator.py.
• The import statement has transitioned:
  • From:

    from SigProfilerMatrixGenerator.scripts import SigProfilerMatrixGenerator

  • To:

    from SigProfilerMatrixGenerator.scripts import MutationMatrixGenerator

These modifications introduce import errors in SigProfilerSimulator (line 20, line 24).