We have collected a large number of genome-scale metabolic models (GEMs) and investigated the aggregated results of running the models through memote's test suite. The models can be found separately at https://github.com/biosustain/memote-meta-models.
The goals of this study were:
- Investigate the collective performance of those models.
- Use those insights to calibrate how memote calculates the final test score.
N.B.: We looked at distributions of test metrics because we are not interested in shaming individual model authors but we are interested in general trends and the current overall state of GEMs.
The easiest way to set up the dependencies for this project is to use the Makefile.
make requirements
make jupyter
This will install all Python requirements and configure the Jupyter notebook
extensions. R dependencies are handled by a specialized Docker image
midnighter/knit-memote:3.6.1.
You can directly use that image in order to ensure reproducibility. If, for some
reason, you wish to build the image yourself, you can do so with the command
make build.
The main work can be performed via the make command make plot or for
more fine grained control via the command line interface exposed by
./cli.py and the relevant R scripts.
For comments and questions get in touch via
- The memote gitter chatroom or
- GitHub issues
- Copyright (c) 2017-2019, Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark.
- Free software: Apache Software License 2.0
- All results found in
data/are available under the CC BY 4.0 license