DIpartite

DIpartite is a tool for detecting bipartite motif based on dinucleotide weight matrix.

DIpartite predicts transcription factor binding sites (TFBSs) based on PWM or DWM. Bipartite motif is defined as two conserved blocks separated by variable gaps.

---

Getting started

git clone https://github.com/Mohammad-Vahed/DIpartite
cd DIpartite
make

---

Usage

<<PWM base prediction>>
## For one block motif
./DIpartite -i <fasta> -n 1 -p 2 -m 6 -M 0 -g 0 -G 0 -o <output>

## For two block motif
./DIpartite -i <fasta> -n 2 -p 2 -m 6 -M 0 -g 0 -G 0 -o <output>


<<DWM base prediction>>
## For one block motif
./DIpartite -i <fasta> -n 1 -p 2 -m 6 -M 0 -g 0 -G 0 -o <output>

## For two block motif
./DIpartite -i <fasta> -n 2 -p 2 -m 6 -M 8 -g 11 -G 16 -o <output>

---

Arguments

-i input file
-m left motif width (default 6)
-M right motif width (default 6)
-g min gap between two motif blocks (default 0)
-G max gap between two motif blocks (default 0)
-t number of times trying to repeat process to find best motif (default 30)
-o output file (default output.txt)
-f 1 for fasta file, or 2 for text file (no header) (default 1)
-p 1 for the given strand, or 2 for both the given and reverse complement strands (default 1)
-n 1 for PWM, or 2 for DWM (default 1)
-s 1 for one occurrence motif site per sequence, or 2 for any number of repetitions or 3 Zero or One Occurrence per Sequence (zoops) (default 1)

---

Reference

1. Mohammad V, Ishihara J, Takahashi H. 2018 in submitting