Rosetta-learn is a recommender system for protein sequence optimization using Rosetta. Users who want to model their high-througput protein sequencing results and associated Rosetta metrics can do so easily with Rosetta-learn.
Rosetta-learn builds and tunes an optimal deep neural network (DNN) to model protein data. Using this model, Rosetta-learn recommends an optimized sequence - predicted to maximize experimental output.
pip install rosetta-learn
Rosetta-learn requires an input xlsx file of protein sequencing data and their respecitve Rosetta metrics with the following structure:
Example:
| Sequences | Rosetta Metric 1 | Rosetta Metric 2 | ... | Output |
|---|---|---|---|---|
| actgactg ... | 12 | 4 | ... | 3 |
| actgactg ... | 3 | 8.3 | ... | 5 |
To generate a new model using the command line interface:
rosetta-learn input.xlsx
To retrain a previously generated model using the command line interface:
rosetta-learn input.xlsx -m model.h5