pycoMeth

CGI_Finder

Simple method to find putative CpG islands in DNA sequences by using a sliding window and merging overlapping windows satisfying the CpG island definition. Results can be saved in bed and tsv format.

In order for a CpG island to be called, it must fulfill three criteria:

• Sufficient lenght. Threshold set my -w argument, default 0.
• Sufficient GC content computed as (N_C + N_G) / N. Threshold set by -c argument, default 0.5.
• Sufficient normalized CpG ratio, computed as (N_CG * N) / (N_C * N_G). Threshold set by -r argument, default 0.6.

Usage

pycometh CGI_Finder [-h] -f REF_FASTA_FN [-b OUTPUT_BED_FN] [-t OUTPUT_TSV_FN] [-m MERGE_GAP] 
                    [-w MIN_WIN_LEN] [-c MIN_CG_FREQ] [-r MIN_OBS_CG_RATIO] [-v] [-q] [-p]

Arguments

• -f REF_FASTA_FN: Filename to genome reference in FASTA format
• -b OUTPUT_BED_FN: Output file in BED format. Optional.
• -t OUTPUT_TSV_FN: Output file in tab-delimited format. Optional.
• -m MERGE_GAP: minimum distance between CpG islands. CpG islands closer than this will be merged to one. Optional. Default: 0
• -w MIN_WIN_LEN: Minimum CpG island length. Optional. Default: 0
• -c MIN_CG_FREQ: Minimum GC frequency. Optional. Default: 0.5
• -r MIN_OBS_CG_RATIO: Minimum normalized CpG ratio. Optional. Default: 0.6
• -v: verbose mode
• -q: quiet mode
• -p: display progress bar

Output files

Tabulated TSV file

This tabulated file contains the following fields for each CpG island found:

• chromosome / start / end : Genomic coordinates
• length: Length of the interval
• num_CpG: Number of CpGs found
• CG_freq: G+C nucleotide frequency
• obs_exp_freq: Observed versus expected CpG frequency

BED file

Minimal standard genomic BED3 format listing the coordinates of putative CpG islands.

This site is open source. Improve this page.