neutral_sequence_generator

Creates a DNA sequence that has motifs distributed at a specified frequency seperated by a neutral sequence that purposefully does not contain the motifs. This program uses SiteOut which was developed by researchers at Harvard University.

code

• Neutral_Sequence_Generator_Notebook.ipynb: Looks at the structure and functions of the program
• Neutral_Sequence_Generator.py: This script program that does the work producing neutral sequences with the TFBS. To use call the sequence_generator() function with your parameters.
• SiteOut.py: This is the algorithm used to produce the neutral sequence.

example_outputs

• /my_example.faa: This is the fasta file that contains the neutral sequences
• /Sequences.txt & /Motifs.txt: These files are used by siteout.py to produce the sequences
• motif_pfm/: This folder contains four transcription factor binding site position frequency matrices that are most involved in drosophila development. To create your own motif_pfm folder download pfm from JASPAR database, I had to modify these by deleting all words and letters before the position frequencies.

control_seqs

-Motifs used (in order): hb, bcd, cad, gt.

There are three different types of control sequences:

• There are 500 random sequence files that were generated with no TFBS motifs. These are the neutral_XXX.fa files.

• There are 2000 sequence files that were generated with a motif probability of .25 for each motif. Meaning there are 500 of each of four motifs, where that motif is at a probability of .25 and the rest are at 0. These are named freq_0.25XYYY.fa Where the 0.25 represents the probability. X represents which motif it is in the list of motifs and YYY represents which number of the 500 it is.

• There are a large number of sequences that had systematically generated motif probabilities. The probability of each motif was able to be controlled independently. Each motif was cycled through a probability ranging from .01-.1. These are named freqrange_X_Y_W_Z.fa Where each letter represents the percent probability that the motif appears in the sequence.

References

Estrada J, Ruiz-Herrero T, Scholes C, Wunderlich Z, DePace AH (2016) SiteOut: An Online Tool to Design Binding Site-Free DNA Sequences. PLoS ONE 11(3): e0151740. https://doi.org/10.1371/journal.pone.0151740