In this capstone project course, we'll compare genome sequences of COVID-19 mutations to identify potential areas a drug therapy can look to target. The first step in drug discovery involves identifying target subsequences of theirs genome to target. We'll start by comparing the genomes of virus mutations to look for similarities. Then, we'll perform PCA to cut down our number of dimensions and identify the most common features. Next, we'll use K-means clustering in Python to find the optimal number of groups and trace the lineage of the virus. Finally, we'll predict similarity between the sequences and use this to pick a target subsequence. Throughout the course, each section will consist of a programming assignment coupled with a guide video and helpful hints. By the end, you'll be well on your way to discovering ways to combat disease with genome sequencing.
Overview
Syllabus
- Comparing Genome Sequences
- In this module, we'll start to get familiar with our dataset by performing some basic EDA and comparing genome sequences. By analyzing the mutations of the COVID-19 virus, we'll be able to identify some common properties of the genome that our drug should look to target.
- Principal Component Analysis on Genome Sequences
- In this module, we'll continue to work with out genome sequence data - using PCA to identify groups and delicate the most important features. After reducing the number of dimensions in the dataset, we'll be able to use K-means to form clusters and visualize the different areas in 2-D space.
- Feature Analysis using K-Means Clustering
- In this module, we'll cluster the genome sequences using the K-means algorithm. We'll optimize the number of clusters by comparing silhouette scores across a wide variety of inputs to identify the greatest drop-off. Finally, we'll set ourselves up to using prediction pipelines to predict bit scores and drug therapies in the last module.
- Predicting Bit Score to Find Sequence Matches
- In this module, we'll test a variety of regressors to see which one performs best in predicting bit scores for each genome sequence. Then, we'll use our chosen model to find the genome equines that are most closely related and trace out a possible subsequence to target with a combative drug.
Taught by
Rajvir Dua and Neelesh Tiruviluamala
