Transcription Factor Binding Prediction from ATAC-seq and scATAC-seq with Deep Neural Networks
peaks will take a maxATAC prediction
.bw signal track and call intervals of TFBS that meet a given confidence threshold. These TFBS intervals will be output as BED intervals that can be visualized and used for downstream analysis.
The peaks function takes as input a bigwig signal track and will output bins that are above a given threshold.
maxatac peaks -i GM12878_CTCF.bw -o ./peaks -bin 32 -cutoff_file ARID3A_validationPerformance_vs_thresholdCalibration.tsv
The input maxATAC bigwig file.
The cutoff file provided in /data/models that corresponds to the average validation performance metrics for the TF model.
Note on abbreviations:
The cutoff type (i.e.
log2FC). Default: F1.
The cutoff value for the cutoff type provided. Note precision, recall, and F1-scores range 0-1, while better-than-random log2FC scores range from 0 to infinity. Example: .7
The prefix to use for the output file name.
The bin size (TFBS interval length) used for calling peaks. Default: 32 bp, based on the benchmarking intervals predictions. 32 bp, the resolution of the maxATAC models, is also a good option.
The path to the output directory to write the bed.
The chromosomes to limit peak calling to. Default: Autosomal chromosomes that are used in training and evaluation.