A Summer of Internship: Week 4-5 at Adebali Labs

A Summer of Internship: Week 4-5 at Adebali Labs

Week 4: Plotting Genetics on a Straight Line πŸ“ˆ

Hey blog family! Week 4 at Adebali Labs marked the intersection of genetics and mathematics, with linear regression taking center stage. Let's break it down!

Predicting Repair Metrics Through Linear Regression πŸ§¬βž‘οΈπŸ“Š

Linear regression, often used to predict a dependent variable based on one or more independent variables, became our tool to make sense of repair metrics in the transcribed strands. Our main players? The CSB and XPC mutants.

Using Linear Regression

The beauty of linear regression is its simplicity and power. Through it, we were able to generate coefficients – numbers that transformed a strand's mutation into tangible repair data. Imagine turning a gene's unique quirks into a predictable, numerical value. That's precisely what we achieved this week!

Using Random Forest

Wrapping up Week 4, we've successfully incorporated mathematical models to predict genetic repairs, bridging the gap between numbers and nucleotides. Onward to Week 5, where the journey took an even deeper twist!

Week 5: Fusing Genes and Fine-tuning Formulas 🧩

Hello again! The journey at Adebali Labs moved forward in Week 5, with datasets merging and math formulas being refined. Dive in with me!

Merging Datasets: The Big Picture πŸ–ΌοΈ

First up, we combined the datasets of transcribed strand (TS) and non-transcribed strand (NTS) to create a more comprehensive view. Think of it as piecing together a jigsaw puzzle to get the complete picture.

With this unified data, the complexity grew, revealing relationships that were initially hidden. This newfound depth meant that our linear regression models and coefficients from Week 4 needed a relook.

Refined Coefficients, Sharper Insights 🎯

The quest this week was iterative. With every cycle of tweaking and validation, our understanding grew sharper. It was like tuning a musical instrument until the note was pitch perfect.

By the end of Week 5, we not only had a holistic dataset but also refined coefficients that allowed us to understand genetic repair with even greater accuracy. The complexity of merging datasets and refining our approach provided a solid base for the project's future phases.

In essence, Weeks 4 and 5 have been all about embracing mathematics to unravel genetic mysteries. With each week, the roadmap to understanding our primary research question becomes clearer. Stay tuned as the journey continues to unfold in the fascinating world of genomics!

Till the next post – keep exploringβœ¨πŸ”