PRS_converter

Convert Tombo per-read statistics files into CSV files.

Usage

Suppose you have a Tombo per-read statistics file at /path/to/prs/file, and you want to transform it into a CSV file at /path/to/future/csv/file.csv. Then run the following code at your terminal. (Don't type the $.)

$ python /path/to/PRS_converter.py /path/to/prs/file /path/to/future/csv/file.csv

(If you are in the same directory as this readme file, use PRS_converter.py as /path/to/PRS_converter.py.)

PRS_converter will not delete the per-read statistics file.

Output format

The output CSV file will have a row for every read and a column for every nucleotide position. Rows are labeled by labeled by their read IDs, and columns are labeled by numbers representing nucleotide positions. (These labels come from the "read_id" and "pos" fields of Tombo's per-read statistics file. "pos" seems to correspond with nuleotide position on the reference genome.)

Troubleshooting

Make sure Python is installed (or that a Python module is loaded). This script was written for Python 2 but it should work with Python 3 as well. You can check Python's version (and whether it's installed) by typing python --version at the terminal.

To see available command-line options for PRS_converter.py, run

$ python /path/to/PRS_converter.py -h

at your terminal.

Notes

• This script was inadequately tested. Be careful.
• Per-read statistics files are produced by Tombo's tombo detect_modifications command. If tombo detect_modifications is fed the --processes or --multiprocess-region-size options then this script will not work right: it will only produce a CSV file containing per-read statistics for part of your genome.
• The output CSV may be smaller than the per-read statistics file, because the per-read statistics file stores data in an inefficient way.
• Ignore the Jupyter notebook(s).