Convolutional Neural Network for prediction of smallRNA:target site binding based on given smallRNA and target sequence.
The user-friendly web application for miRNA and tRNA target predictions https://evaklimentova.github.io/smallRNA_binding/
Using Git:
git clone https://github.com/evaklimentova/smallRNA_binding.git
or
git clone git@github.com:evaklimentova/smallRNA_binding.git
It is implemented in python using Keras and Tensorflow backend.
Required:
- python, recommended version 3.7
- Keras 2.7.0
- tensorflow 2.7.0
- pandas
- numpy
# create a virtual environment:
python -m venv env
# activate it and install the necessary libraries.
source env/bin/activate
pip install -r requirements.txt
To make the predictions, choose one of the miRNA (Models/model_miRNA.h5), tRNA (Models/model_tRNA.h5) or yRNA (Models/model_yRNA.h5) models. Required input is a tsv file with multiple potential small RNA - target pairs consisting of first column containing small RNA sequence (idealy around 20 bp long) and second column containing exactly 50 bp long target sequence.
# if you are not actively sourcing from the previously created virtualenv:
source venv/bin/activate
# run the prediction
./predict.py --input <input_file> --output <output_file> --model <Models/model_miRNA.h5>