MEX-Archipelago data

This repository contains preprocessed Codebook-MEX data suitable for training and testing advanced models of transcription factor binding sites, see README for details.These data are derived from the ChIP-Seq and genomic HT-SELEX datasets of the primary MEX data release (ZENODO doi:10.5281/zenodo.8327372). Extra: ARCHIPELAGO_models.zip: the ArChIPelago models trained on genomic HT-SELEX and ChIP-Seq data for 139 TFs. Please refer to 5_Archipelago_sav_models_loading_and_scoring.ipynb in the ArChIPelago GitHub repo (https://github.com/autosome-ru/MEX-ArChIPelago) for details on using the model for scanning arbitrary DNA sequences. * Latest update: Archipelago models and underlying data were removed for three TFs (TIGD4, TTF1, ZBED5), which were partially affected by mislabeled samples in the original upload.