FLIGHTED: Inferring Fitness Landscapes from Noisy High-Throughput Experimental Data (Part 1)

Data for FLIGHTED (Inferring Fitness Landscapes from Noisy High-Throughput Experimental Data). This data contains the model weights for FLIGHTED-Selection and FLIGHTED-DHARMA, the training data for both, and fits on the GB1 landscape. It does not contain anything related to the TEV protease landscape.

The data is arranged in the following folders:

DHARMA_Input: contains the input for the DHARMA models, with the canvas sequence, the DHARMA reads, and the FACS data.

DHARMA_Models: contains the weights, hyperparameters, and model training history for the FLIGHTED-DHARMA model.

Fitness_Landscapes: the GB1 landscape, with and without FLIGHTED, as well as splits published by FLIP.

Landscape_Models: models trained on the GB1 landscape with and without FLIGHTED under the various FLIP splits. Each model folder contains hyperparameters, training history, and predictions on the test set which can be used to evaluate model performance. Raw model parameters are not provided for fine-tuned models due to size; contact us if you want them.

FLIGHTED_Selection: contains the weights, hyperparameters, and model training history for the FLIGHTED-Selection model.

Selection_Simulations: contains the simulated training data for the FLIGHTED-Selection model.