ADITYA ELAYAVALLI

Driven by a passion for computational biology and a track record of success at Unravel Biosciences, I excel in developing algorithms and models to advance drug discovery. My expertise in Python, TensorFlow, and collaborative problem-solving has led to significant improvements in drug prediction engines, underscored by a high-accuracy CNN model for breast cancer diagnosis.

GENOMICS PROJECT, Boston, MA, Northeastern University, 10/23,

• Given a Genomic sequence, developed Python code to find the number of ORFs (Open Reading Frame) being produced, and ensure they have a threshold score over 7.25. As part of this project I utilized BioPython and Matplotlib packages. The detailed findings were later incorporated in a presentation.
• Created a Phylogenetic tree from a distance matrix using the UPGMA (unweighted pair group method with arithmetic mean) approach. This enabled identification of similarities between various species as part of the dataset.
• Created a Python script to analyze genetic sequences, specifically the codons, and determine if they are coding for a protein or not, using a coding model matrix and a non-coding model matrix. This project involved producing a ROC curve, visualizing the performance of the binary classifier showing the distinction between coding and non-coding sequences. This project was completed using NumPy, Matplotlib and Sklearn.metrics packages.

MACHINE LEARNING PROJECT, Boston, MA, Northeastern University, 10/23 - 12/23, 

• Developed three machine learning classification models (Naive Bayes, Logistic Regression, and Decision Tree) to predict diabetes diagnosis. Followed the CRISP-DM process, including data sourcing, cleaning, transformation, imputation, validation strategy, model tuning, and evaluation.

United States of America