Skip to content

Genedisco work for the `pycd` team

License

Apache-2.0, MIT licenses found

Licenses found

Apache-2.0
LICENSE.txt
MIT
LICENSE_MIT.md
Notifications You must be signed in to change notification settings

chrisemezue/genedisco-pycd

Repository files navigation

GeneDisco competition 1st place submission (team pycd)

image

We used Random Network Distilation model (code snippet) for our final submission. Please refer to the paper to learn more about the method.

GeneDisco: A benchmark for active learning in drug discovery

Python version Library version

In vitro cellular experimentation with genetic interventions, using for example CRISPR technologies, is an essential step in early-stage drug discovery and target validation that serves to assess initial hypotheses about causal associations between biological mechanisms and disease pathologies. With billions of potential hypotheses to test, the experimental design space for in vitro genetic experiments is extremely vast, and the available experimental capacity - even at the largest research institutions in the world - pales in relation to the size of this biological hypothesis space.

GeneDisco (published at ICLR-22) is a benchmark suite for evaluating active learning algorithms for experimental design in drug discovery. GeneDisco contains a curated set of multiple publicly available experimental data sets as well as open-source i mplementations of state-of-the-art active learning policies for experimental design and exploration.

GeneDisco ICLR-22 Challenge

Visualization

Please note that it is possible to open genedisco-pycd/genedisco/visualization/visualization_2.ipynb, even though it's too large to view it on github. To do that, open the notebook as a raw file, copy its content, and save it on your local machine as an .ipynb file. The notebook contains some of our latest visualization comparisons that we used to choose the best performing acquisition function among those we tried.

License

GeneDisco benchmark is licensed under Apache License.

Contributions from pycd team are licensed under MIT License:

genedisco-pycd/genedisco/active_learning_methods/acquisition_functions/core_set.py
genedisco-pycd/genedisco/active_learning_methods/acquisition_functions/core_set2.py
genedisco-pycd/genedisco/active_learning_methods/acquisition_functions/core_setUMAP.py 
genedisco-pycd/genedisco/active_learning_methods/acquisition_functions/ensemble_rnd.py 
genedisco-pycd/genedisco/active_learning_methods/acquisition_functions/rnd.py
genedisco-pycd/genedisco/active_learning_methods/acquisition_functions/rnd_05.py
genedisco-pycd/genedisco/active_learning_methods/acquisition_functions/uncertainty_acquisition.py
genedisco-pycd/genedisco/active_learning_methods/acquisition_functions/uncertainty_acquisition_03.py
genedisco-pycd/genedisco/active_learning_methods/acquisition_functions/uncertainty_acquisition_05.py
genedisco-pycd/genedisco/active_learning_methods/acquisition_functions/uncertainty_acquisition_07.py
genedisco-pycd/genedisco/active_learning_methods/acquisition_functions/uncertainty_acquisition_10.py
genedisco-pycd/genedisco/visualization/visualization_2.ipynb
genedisco-pycd/genedisco/visualization/viz.py
genedisco-pycd/genedisco/visualization/viz_utils.py

PYCD Team

About

Genedisco work for the `pycd` team

Resources

License

Apache-2.0, MIT licenses found

Licenses found

Apache-2.0
LICENSE.txt
MIT
LICENSE_MIT.md

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages