StrucTFactor

Transcription factor prediction using protein 3D secondary structures

We refer to our GitHub for more details: https://github.com/lieboldj/StrucTFactor

Contact Information: Khalique Newaz
E-mail: khalique.newaz@uni-hamburg.de

Transcription factors (TFs) are DNA-binding proteins that regulate gene expression. Traditional methods predict a protein as a TF if the protein contains any DNA-binding domains (DBDs) of known TFs. However, this approach fails to identify a novel TF that does not contain any known DBDs. Recently proposed TF prediction methods do not rely on DBDs. Such methods use features of protein sequences to train a machine learning model, and then use the trained model to predict whether a protein is a TF or not. Because the 3-dimensional (3D) structure of a protein captures more information than its sequence, using 3D protein structures will likely allow for more accurate prediction of novel TFs.

Overview of our TF prediction method: StrucTFactor

Software / Hardware prerequisites

Before you begin, ensure that you have the following prerequisites installed on your system:

Anaconda or Miniconda
pip package manager
python >= 3.6.12
Models have been trained on a single Tesla T4 GPU.

Installation Steps

Given CUDA version 12.2, the following installation guide was tested:

1. Clone the repository:
   git clone https://github.com/lieboldj/StrucTFactor.git

2. Create a conda environment:
   conda env create -f stf.yml
   conda activate stf

If you want to benchmark with DeepReg, please install tensorflow.

Predict TF/non-TF for a pdb file

Given a pdb file, you run the following to predict whether this protein is a TF:

cd strucTFactor
python predictTFwithStrucTFactor.py -i <pdb_file>

The result will be printed in your terminal and written in output.csv.

Please cite: Liebold J., Neuhaus F., Geiser J., Kurtz S., Baumbach J., Newaz K. (2024). Transcription factor prediction using protein 3D secondary structures.