bbgMod is a computational tool that employs nanopore data to extract modified positions. The model employs two sets of reads, one from the untreated/control sample and one from the treated sample. Statistical comparison of both sets is carried out so as to extract a curated list of modified positions
First download bbgMod from the github repository:
git clone add/final/github/path
We highly recommend to use a virtual environment for the installation and employment of bbgMod:
Option 1:
pip install -e .
In order to install tombo:
conda install -c bioconda ont-tombo
Detect modifications:
bbgMod detect-modifications -f5 path/to/control/fast5/reads/ -af5 path/to/alternate/fast5/reads/ -cpu 56 -o path/output
Example in the cluster:
pwd: /workspace/projects/nanopore
python analysis_libs/stats.py -f5 /workspace/datasets/nanopore/experiment_3/X4_2_FAL40053/fast5/pre_resquiggle_merge/untreated/reads/ -af5 /workspace/datasets/nanopore/experiment_3/X4_2_FAL40053/fast5/pre_resquiggle_merge/cisplatin/reads/ -o test/test_modifications/
Run logistic regression on errors analysis:
bbgMod error-analysis -te path/to/extracted/errors/treated -ue path/to/extracted/errors/untreated -cpu 56 -o path/output
Location of error features in the cluster:
/workspace/projects/nanopore/stockholm/EpiNano/novoa_features
Location of data sets of our experiments in the cluster (further explanation needed):
/workspace/datasets/nanopore