MSAffect is a computational pipeline to evaluate the robustness of AlphaFold2 protein structure predictions under adversarial MSA perturbations to identify structural sensitivity and confidence shifts in neural-network-based folding.

Implementation of two perturbation functions for Knapsack Problem

Python implementation of a data perturbation method to determine relevant features for NN learning

A reproducible computational pipeline for processing and analyzing single-cell RNA-seq data with CRISPR perturbations, designed for the Virtual Cell Challenge 2025. Features automated quality control, normalization, class balancing, and batch integration using Snakemake.

Exploring perturbation stategies to improve LIME explanations stability.

Dataset of SOSAA model trajectory runs

M0dular Action System & Interface, Custom AI/AGENT Framework (made by a non-coder)

A demonstration of perturbation of data

A Python simulator for modeling gene expression perturbations using a recursive, depth-limited path summation model.

Presentation given to Drexel XAI seminar on "Interpretable Explanations of Black Boxes by Meaningful Perturbation"

Symbolic calculation of the (van Vleck) Floquet recursion formula

Assessing consciousness by perturbing a dynamic mean-field whole-brain model fitted to empirical neuroimaging data

GUI for rust-fractal-core, a mandelbrot fractal visualizer featuring perturbation based iteration methods, series approximation and multithreading.

Second Year project (Module: Artificial Intelligence Methods), focusing on implementing and evaluating AI Optimisation methods.

DataArmor is a cutting-edge tool focused on safeguarding privacy in today's data-driven world using K-anonymity L-diversity and t-closeness privacy model. As the sharing of personal and microdata grows, ensuring the protection of individual identities during data publication and analysis becomes essential.

compact: co-expression module perturbation analysis

Interpretable Explanations of Black Boxes by Meaningful Perturbation Pytorch

Perturbation toolbox to solve DSGE models with Matlab

Mandelbrot Generator based on perturbation and series approximation

Symbolic Perturbation Theory (SymPT) is a Python package for symbolic perturbative transformations on quantum systems. SymPT helps compute effective Hamiltonians for both time-independent and time-dependent systems at both operator and matrix level.