Metagenomics Deep Learning

Introduction

This is the reference material for the "Deep Learning in Metagenomics" workshop. By the end of workshop, you will be able to understand what is deep learning and how to apply to metagenomics taxanomic classification.

Click here for "Metagenomics - Machine Learning" workshop

Environment Setup

We will be using Python, Jupyter notebook and command line tools for this workshop.

If you are using Windows, you can install WSL2 and use Ubuntu. If you are using Mac, you can use the terminal.

Install anaconda
Create new environment
Install required packages

$ wget https://repo.anaconda.com/archive/Anaconda3-2024.06-1-Linux-x86_64.sh

$ bash Anaconda3-2021.05-Linux-x86_64.sh

$ conda create -n mgdl -c fkretschmer bertax

$ conda activate mgdl

$ pip install keras-bert==0.86.0 numpy==1.23.0

$ mkdir -p workshop; cd workshop

$ git clone https://github.com/rnajena/bertax

$ pip install -e bertax

$ cd bertax

$ cd bertax/resources

$ wget https://github.com/rnajena/bertax/releases/latest/download/big_trainingset_all_fix_classes_selection.h5

Machine Learning Notebooks

https://github.com/ChillarAnand/avilpage.com/tree/master/mg_workshop

We will build our own neural network to classify MNIST database.

This will help use understand how MLP works.

Then we will look into Transformers(these are used in ChatGPT and other LLMs).

Finally we will run bertax model to do metagenomics taxanomic classification.

Resources

Feedback

Please take a moment to provide your feedback here