Neural MSA
Scaling a neural network that solves DNA MSA for real-world use.
I’m a B.Sc. student in Psychology at McGill University. My research focuses on building transformer-based models for cognitive tasks and DNA multiple sequence alignment, as well as developing open-source pipelines for behavioral data modeling. I’m especially interested in aligning artificial neural networks with human brain activity to gain insight into cognition. My long-term goal is to contribute to clinical psychology by using machine learning and neural network models to study how pathologies develop in the brain and at the genetic level.
Modeling brain activity for different tasks to understand how the brain solves them; comparing control and experimental groups.
Improving heuristic/hand-crafted pipelines by leveraging neural networks to enhance measurement quality.
Making computational research more accessible through reusable packages.
Lucas Gomez, Aziz Ktari, Hao Yuan, Bai Pouya Bashivan (2025). Data on the Brain and Mind (NeurIPS Workshop). [Accepted] · Link
Research ways to improve Multiple sequence alignment tools using ANNs (2025).
Scaling a neural network that solves DNA MSA for real-world use.
Aligning model activations with PFC fMRI; encode/delay phase analyses.
Pluggable Pipeline for RL/choice/bandit agents that handles simulation, recovery, fitting and plotting.
When I come up with an idea for an app I’d find useful, I enjoy coding it for myself. I find it fun, and it lets me integrate all the features I want that I might not find combined, or even available, in other apps.
I play basketball and soccer, and I’m also part of a dodgeball team. I enjoy watching MMA and soccer with friends.
Email: aziz.ktari@mail.mcgill.ca
Location: Montréal, Canada
I’m open to collaborations and mentorship discussions on applying artificial neural networks to study clinical pathologies, from genetic risk factors and physiological markers to brain activity patterns and patient-specific differences. Always happy to connect with researchers exploring how machine learning can reveal early indicators and mechanisms of mental health conditions.