Predicting-gene-expression-using-millions-of-random-promoter-sequences

One contribution from Team: DrAshokAndFriends to random-promoter-dream-challenge-2022.

Includes:

1. Final report.docx: Documentation of our contribution to this challenge. More details about the competition could be found at: https://www.synapse.org/#!Synapse:syn28469146/wiki/617563
2. Code.py: The entire code for preprocessing the data, building the model, training the model, and evaluating the model. Requires file 'train_sequences_exp.txt' with only the expression values in the training data (one value per line), to generate the sample weights.
3. TrainedModel.h5: The model trained as per the documentation details, which could be loaded to make new predictions
4. SubmissionFile.txt: Tab-delimited file of sequences and predicted expression values on the test set.