This repository contains scripts for preprocessing Nanopore DNA sequencing data for training and validating BiGRU Recurrent Nerual Neatworks for detecting BrdU, Jupyter Notebooks for model training, and models I-III
- BiGRU RNN Models I-III: three Kera BiGRU RNN models trained with incremental data
- dataset_I_scripts: scripts used to processing Dataset I, 5-mer standard references generated by using the barcoded split pool synthesis method, BSBP, sequenced on Nanopore R9.4.4 flow cells.
- dataset_II_scripts: scripts used to processing Dataset II
- dataset_III_scripts: scripts used to processing Dataset III
- dataset_IV_scripts: scripts used to processing Dataset IV
- Jupyter.notebooks.for.model.training.zip: Jupyter notebooks for training Models I-III