Illustration to running workflow of web-server and standalone program

Description

PTFSpot: Deep co-learning on transcription factors and their binding regions attains impeccable universality in plants.

Running the standalone program

Requirements

1. Python3
2. Numpy
3. keras
4. tensorflow
5. plotly
6. pandas
7. bayesian-optimization
8. bedops
9. Bedtools
10. Alphafold2 generated PDB files

To build model implementing hyperparameter tuning

Example: python3 hyper_param.py file_for_tuning

Input file description

file_for_tuning = file containing label (0/1) and sequence (positive and negative instances). All in one line separated by tab for a single instance.

Label	Sequence
1	TGATAAACAAAGTGTGTAACATCACCTCATCTACATGTGTGATTTTTTTTTTGAATATAGACAACTTTTTAGTCAGAGTAGTGAGTATAGTGAGTTTCTGTAGAGAAGCTCATCTTAGAATTATTCATGTATTCCACTACTAAAATGTATTCCACTACT
0	AGATCTACAAGAGAAGATAAGTTTGAGGCAAATTCGAGATCTGGAAGCTGGTTTTCTCTTTACAAATAACACTAACCCTACCATCAAATCAAGAAAGGAGGCTTTGAACAAATAGCTTGATTGAAGTATGAAGTGGCTCGGTGGGCGACGATGA

Output file description

param.txt = Optimized hyparameters for Transformers.

Name	Value
learning_rate	0.583
activation	relu
activation2	selu
activation3	LeakyRelu
batch_size	40
embed_dim	28
epochs	20
num_heads	14
ff_dim	14
neurons	38
neurons2	12
dropout_rate	0.16
dropout_rate2	0.17
Optimizer	Adadelta

ptfspot.h5 = Hyperparameter optimized trained model for transformer.

Note

Always place your Alphafold2 generated TF pdb file in "pdb" folder (an example is provided).

This project was executed within the Ubuntu Linux platform's Open Source OS environment

Running script

Module: Transformer-DenseNet system (identify TF regions within sequence with the corresponding Alphafold2 generated TF protein structure)

To detect the TF binding region, In parent directory execute following command:

• unzip deeptfactor.zip
• chmod a+x PTFSpot.sh
• sh PTFSpot.sh fasta_file folderpath Alphafold2_generated_PDB_file folder_path_of_PTFSpot
• example: sh PTFSpot.sh genomic_sequence /home/user/fasta ABF2 /home/user/PTFSpot

Output description

TFbinding regions detection module gives output in following format

ABF2_genomic_sequence.txt = TF binding regions result (Final result: ID, Start, End).

Sequence ID	Start	End
seq1	30	190
seq2	10	170
seq3	21	151
seq4	73	233

To generate line plot, execute the following command:

python3 make-plot.py seq1.csv (generated plots are interactive)

An example of generated plot:

Python script to build model implementing hyperparameter tuning Python version 3.6 or higher Version 1.23.5 Version 2.13.1 Version 2.13.1 Version 5.11.0 Version 1.5.0 Version 1.2.0 sudo apt-get install bedops -y sudo apt-get install bedtools -y Alphafold2: https://alphafold.ebi.ac.uk/ Ubuntu: https://ubuntu.com/ python script to generate plots filename Complete execution shell script Name of the fasta sequence containing file Complete path of the folder which contains fasta file Name of the pdb file generated from Alphafold2 Complete path of the folder containing scripts of PTFSpot >seq
TGATAAACAAAGTGTGTAACATCACCTCATCTACATGTGTGATTTTTTTTTTGAATATAGACAACTTTTTAGTCAGAGT
AGTGAGTATAGTGAGTTTCTGTAGAGAAGCTCATCTTAGAATTATTCATGTATTCCACTACTAAAATGTATTCCACTAC
TAGATCTACAAGAGAAGATAAGTTTGAGGCAAATTCGAGATCTGGAAGCTGGTTTTCTCTTTACAAATAACACTAACCC
TACCATCAAATCAAGAAAGGAGGCTTTGAACAAATAGCTTGATTGAAGTATGAAGTGGCTCGGTGGGCGACGATGAGCA Name of the TF pdb file without ".pdb" file_for_tuning = file containing label (0/1) and sequence (positive and negative instances). All in one line separated by tab for a single instance. Click to know more Click to download an example pdb file Click to expand Hyperparameter optimized trained model for transformer.