Introduction

This repository contains code to reproduce the results of the correspondence.

Files

Jupyter notebooks

• jeme-ripple.ipynb: To run JEME and RIPPLE using RIPPLE data with various parameter settings.
• jeme.ipynb: To run JEME and TargetFinder using JEME "random targets" datasets with various parameter settings.
• motifs prediction targetfinder-results.ipynb: To obtain the prediction performances of using Wang et al motifs and shuffled motifs.
• reproducing TF paper-eep.ipynb: To obtain the cross-validation with shuffling and chromosome-split results of TargetFinder on its EE/P datasets.
• reproducing TF paper.ipynb: To obtain the cross-validation with shuffling and chromsome-split results of TargetFinder on its E/P/W datasets.
• shuffle_motif.ipynb: To generate shuffled motifs.

HTML files

• tomtom-shuffled-motifs.html: TomTom results using shuffled motifs against JASPAR 2014 Vertebrates motif database.
• tomtom-wang-motifs.html: TomTom results using Wang et al motifs against JASPAR 2014 Vertebrates motif database.

meme files

• wang_etal_motifs.meme: The motifs from Wang et al in MEME format.
• wang_motifs_all_shuffled.meme: The shuffled Wang et al motifs in MEME format.

targetfinder-master folder

Bash scripts

• count_overlap_eep_same_promoter.sh: Count the number of extended enhancers overlap with each other at different cutoff thresholds and share the same promoter.
• count_overlap_eep.sh: Count the number of extended enhancers overlap with each other at different cutoff thresholds.
• count_overlap.same_promoter.sh: Count the number of windows overlap with each other at different cutoff thresholds and share the same promoter.
• count_overlap.sh: Count the number of windows overlap with each other at different cutoff thresholds.
• generate_pairs_index_eep.sh: Produce the name tuples of samples with overlapping extended-enhancers at different overlapping cutoffs.
• generate_pairs_index.sh: Produce the name tuples of sampels with overlapping windows at different overlapping cutoffs.

Python scripts

• generate_window3.py: Generate window regions as Bed3 format.

Cell line folders

• output-eep: The data of EE/P datasets.

• output-epw: The data of E/P/W datasets.

  • motifs: Contains the files of Wang et al motif occurrence scores in the window regions.

  • motifs_shuffled: Contains the files of occurrence scores of shuffled motifs in the window regions.

    The occurrence scores are calculated by taking the sum of matching scores of all occurrences of the motifs in a window region.