search

calico / basenji

Sequential regulatory activity predictions with deep convolutional neural networks.
Apache License 2.0
397 stars 121 forks source link

Seeking Guidance on bam_cov.py Parameters #185

Closed zhanxiangzong closed 11 months ago

zhanxiangzong commented 11 months ago

Hello, I want to express my sincere gratitude for providing this software. Currently, I'm facing an issue while using bam_cov.py to convert BAM files to BigWig. It seems that using only the default parameters doesn't achieve all the preprocessing steps outlined in your paper (for instance, the -g default in bam_cov.py is set to False). I would greatly appreciate guidance on how to adjust the parameters in bam_cov.py to ensure that the data preprocessing aligns with the standard procedures outlined in your paper. Specifically for CAGE, DNase-seq, ATAC-seq, and ChIP-seq, are there specific parameter settings? Also, are there any parameters that should be modified based on different species? Thank you very much for your time and support.

davek44 commented 11 months ago

Here are some rough recommendations, but note that most of these are chosen somewhat subjectively. I can't think of a reason to modify them across different species.

CAGE: bam_cov.py --clip_multi 12 -m 3 -s 4 --smooth_out 2 --strand DNase/ATAC/ChIP: bam_cov.py --clip_multi 12 -g -m 3 -s 16 --smooth_out 8 For ATAC, add -v 4 -w 5 For ChIP, add -c

zhanxiangzong commented 11 months ago

Thank you very much for your help.