This repository contains the code developed for the paper "Deep protein representations enable recombinant protein expression prediction".
Most of the code is originally from the UniRep repository, but has been modified to suit this project. The script is src/unirep_formatter.py
.
Two scripts contains the code for training classifiers to predict protein expression:
src/train.py
: Trains SVM, LR and RF classifierssrc/train_ann.py
: Train ANN classifiers
The notebook Figures_and_tables.ipynb
show how to generate the figures and tables of the paper. Specifically Figure 2, Figure 3, Table 1 and supplementary material.