CS234: HMM based DNA methylation states predictor using BW algorithm and comparison with DeepCpG

Searching project topic (Jan 25, 2021)

• "DeepCpG accurate prediction of single-cell DNA methylation states using deep learning", Genome Biology, 2017

• "Using neural networks for reducing the dimensions of single-cell RNA-Seq data", NAR, 2017.

• Gene expression inference with deep learning, Bioinformatics, 2016.

• Improving protein disorder prediction by deep bidirectional long short-term memory recurrent neural networks, Bioinformatics, 2016.

• "deepTarget: End-to-end Learning Framework for microRNA Target Prediction using Deep Recurrent Neural Networks", BCB 2016.

• A deep learning framework for modeling structural features of RNA-binding protein targets, Nucleic Acids Research, 2016.

Finalized project topic (Feb 20, 2021)

Implementing an HMM based DNA methylation states predictor using Baum-Welch algorithm and comparing its performance with DeepCpG.

Implementation specifications (Feb 22, 2021)

• Programming language: Python

• Environment: Ubuntu 18.04 lts 64 bit

Datasets (Mar 10, 2021)

• Mouse embryonic stem cells profiled using whole-genome single-cell methylation profiling (genomewide bisulfite sequencing or, scBS-seq)

• Human and mouse cells profiled using a reduced representation protocol (scRRBS-seq)

DeepCpG Architecture (Mar 10, 2021)

Complete Work flow (Mar 15, 2021)

• Please have a look at the short documentation of the project.

• GitHub link to the project will be available soon.