Matthew Masters

Auto Encoding Molecular Conformations:
PyTorch implementation of molecular conformation autoencoder described in Winder 2021.

The first automated machine learning tool specifically for graph data. Optimizes and ensembles several traditional and deep learning methods.

Advent of Code 2020:
Daily programming challenges

Rosalind:
Bioinformatics programming challenges

Mechanism of Action Prediction:
Using gene expression and cell viability data, a deep neural network was used to predict drug MoA

Wheat Head Image Detection:
Object detection pipelines built using FasterRCNN and EfficientDet deep neural networks

Prostate Cancer Grade Assessment:
Deep learning system for detection and classification of prostate cancer

Abstraction and Reasoning Corpus:
Domain-specific language and graph search to solve intelligence tasks

RamaPlot:
Generate Ramachandran plots from PDB structures

Molecular Property Prediction:
Using message passing neural networks to predict NMR scalar couplings

DxViewer:
Web-based DX file viewer. Useful for visualizing densities.

WATsite3.0:
Prediction and Thermodynamic Profiling of Hydration Sites

Project Euler:
Solutions to mathematical challenges written in Python

TrackML Particle Tracking:
Machine learning challenge to isolate particle tracks from CERN detectors

Data Science Bowl 2018:
Mask R-CNN to perform instance segmentation from microscope images