Computer Lab 3: Bioinformatics and Metagenomics

Welcome to the practical session of Module 3, created by lecturer Laura Carroll and Nima Rafati. In this lab, you’ll learn some useful skills from the following Google Colab notebooks.

Notebook 1: Training a Conditional Random Field (CRF) for ribosomally synthesized and post-translationally modified peptides(RiPPs) Detection.

Notebook 2: scRNASeq data analysis workflow.

Access will only be granted after the computer lab session begins.

After opening the notebooks in Google Colab, save a personal copy to your Google Drive: File > Save a copy in Drive. You’ll run and complete your tasks in this copy.

Objective

In this lab, you will:

1. Train a Conditional Random Field (CRF) Model for RiPP Detection: You will use the GECCO tool to train a CRF model specifically for detecting ribosomally synthesized and post-translationally modified peptides (RiPPs) in bacterial genomes. This will involve processing input files, training the model, and evaluating its performance on test data.

2. Conduct Standard scRNASeq Data Analysis Workflow: You will analyze scRNASeq data from patients with significant lung inflammation versus healthy individuals. This workflow includes loading data, quality control, filtering, doublet detection, cell cycle regression, feature selection, data integration, dimensionality reduction, and differential expression analysis.

Important Notes

  • Set Runtime to CPU: We don’t need GPU this time. The Google Colab’s default runtime is CPU, so you can just click the ‘Run’ button to run the code. If you want to ensure your runtime is set to CPU, you can select: Runtime > Change runtime type > Hardware accelerator (set to CPU).

  • Hands-On Learning: This lab emphasizes active participation. You’ll need to carefully study and complete the code snippets provided in the notebooks. Full solutions are not provided, so this is your chance to deepen your understanding.

  • Time Management: It’s OK to upload your work without finishing all tasks during the computer lab. Focus on learning from the exercises you complete, as the primary goal is to understand the key concepts as you work through the notebooks.

Submission Instructions

After completing the notebooks exercises:

1. Export your notebooks from Google Colab:

   - File > Download > Download .ipynb

2. Submit the downloaded files using the provided submission form.

Submission Deadline

You should submit your work by the end of the computer lab session. If you cannot finish it, you can submit it within 24 hours after the lab session ends. So the deadline is Thursday, 12.09.2024, at 17:00 CEST.

If you cannot attend the lab session, you will need to get an approval and also submit your work within 24 hours after the lab session ends.

Next