Manufacturing electronic devices makes negative environmental impacts; alternative approaches involve the use of organic materials. With researchers from University of the West of England and University of Bristol, we developed software for doing extensive statistical analysis and modelling on electrophysiological data collected from laboratory experiments with model fungi. Our full results are published here: https://doi.org/10.1016/j.biosystems.2023.104933 

In collaboration with researchers from University of Oxford, Quadram Institute, Mahdiol University and University of the West of England, we productionised a framework for identifying and classifying virus genomes, based on original work in https://doi.org/10.1186/s40168-018-0422-7. The Maybe Bio Analytics team built in manifold optimizations, wider support, machine learning component refinements, an interactive API, Docker container compatibility, automatic acquisition (scraping) and filtering of virus metadata resource (VMR) files and a different workflow. 

In collaboration with researchers at the University of the West of England, Mayne Bio Analytics created an automation suite for calculating the similarity between metagenome distance (MASH) data and doing clustering therein, using dimensionality reduction and the Louvain method. Methods for graphing and calculating descriptive statistics on output were also implemented. 

Kymographs are a graphical method for representing a dynamical system, by showing how spatial positions therein change over time. Mayne Bio Analytics built scripts for our client, who required automated analysis of microscopy time lapse data, towards showing how oscillating phenomena develop in micro-chemical reactors.