Fingerprints inside COGs strings

This is a mini project code files for "Topics in Bioinformatics 2" course in BGU.

Article: Parikh Mapping-based algorithm for finding gene clusters
Link: http://www.sciencedirect.com/science/article/pii/S1570866703000352

This is an implementation in python2.7.

Before use:

Before use, please install redis.

• For Linux: https://redis.io/download
• For Windows: https://github.com/MSOpenTech/redis/releases

Install modules via pip, if they are not already installed:

pip install redis
pip install ast
pip install datetime

Preprocess:

Before running the program you should run this:

cd <project-folder>
python src/preprocess.py -taxa <path-to-taxa-data-file>
python src/preprocess.py -sigma <path-to-sigma-data-file>
python src/preprocess.py -strings <path-to-strings-data-file>
python src/preprocess.py -cogs <path-to-cogs-info-file>

Available options for preprocess.py:

• -taxa :
  Builds the taxa DB accordingly
• -sigma :
  Builds the sigma DB accordingly
• -strings :
  Builds the strings DB and strains DB accordingly
• -cogs :
  Builds the COGs function DB and the COGs list DB accordingly

Main algorithm:

Now you can run the program (No need to repeat the previous steps on your machine anymore):

cd <project-folder>
python src/run.py <results-directory> <option> <arg>

results-directory is the directory in which the program will save the results file.

• Can be an absolute or a relative path.
• Can be an existing folder or a new directory that the program will create.

Available options for run.py:

• -f :
  Runs algorithm for a specific family, for example: -f bacgroup_Acidobacteria.
• -t :
  Runs algorithm for a specific family type, for example: -t bacgroup.

Postprocess:

For postprocessing:

cd <project-folder>
python src/postprocess.py <results-folder> <family> <options....>

1. results-directory is the directory from which the program will get the results file created when running the run.py program.
   It also will be the directory the postprocess program will save the postprocess results to.
   Can be an absolute or a relative path.
2. family is the specific family name we want to process the results of, for example: bacgroup_Acidobacteria.

Available options for postprocess.py:

• -threshold :
  Runs postprocessing for thresholds of [0.05, 0.1, 0.2, 0.3, 0.5, 0.8], where x in thresholds array is the % of strings of all the strings for this family with the same fingerprint.
• -cogs :
  Runs postprocessing for a specific COGs function list, for example: -cogs ['S','V','V'].
  Finds all fingerprints with those functions that are above threshold (as in the previous option) with the addition of threshold 0 for all the fingerprints w.
• -find :
  Runs postprocessing for a specific COGs list, for example: -find ['0841','0845','3422'].
• -findWithLen :
  Runs postprocessing for a specific COGs list for length constrain, for example: -findWithLen ['0841','0845','3422'] 4.