About me (PhD submitted!!!)

Hey there! I just submitted my PhD at the Alan Turing Institute and Queen Mary University of London. My PhD project focused on Machine Learning appliacations to Precision Medicine. In particular, my project looked at understanding the immune system of trauma patients in order to provide treatment when it gets dysregulated. More broadly, I’m interested in using data-driven methods (both supervised and unsupervised) to disentangle complex biological signals and understand the pathophysiology of immune-driven diseases. I’ve previoulsy worked on multi-omics data integration appraches and representaton learning both in industry and academia. Recently, under the supervision of Prof Barnes and Prof Pennington, I’ve also been involved in developing predictive models for the identification of therapeutic targets/predictive biomarkers to provide diagnosis and treatment to patients affected by these conditions.

Work experience

Part-time ML Engineer at Inflammatix (California, USA): I worked as part-time Machine Learning Engineer on Representation Learning. My work focused on the use of generative models such as Variational AutoEncoders to capture meaningful structure in the data.

Intern at Inflammatix (USA): Interned during summer 2021 at Inflammatix as ML Engineer. My work focused on data harmonisation and the development of a framework for the company.

ML Learning Tutor at Cambridge Spark: Tutored various Machine Learning courses at Cambridge Spark including: “Introduction to Machine Learning”, “Supervised Learning 1 and 2”, “Unsupervised Learning”, “Ensembles”, “Data visualisation”, “Drata Wrangling”, “Python and Pandas”.

ML Summer School at Oxford (2020) OxML Summer School organised by Saïd Business School (Oxford University), AI for Global Goals, Deep Medicine (Oxford University) and CIFAR. 17th - 25th of August

Deep Learning and Reinforcement Learning Summer School at MILA and CIFAR (2020)

News and posters

• August 2023 - New Paper: “The benefits and pitfalls of machine learning for biomarker discovery”
• August 2023 - Submitted PhD!!!
• October 2022 - NeurIPS paper accepted at AI for Science NeurIPS workshop! (conference in Nov-Dec)
• October 2022 - Talk at CSHL on regularisation methods for VAEs (conference in Nov)
• October 2022 - Poster at CSHL “A deep learning approach to integrate different data modalities and improve patient stratification in trauma”
• July 2021 - Poster presentation at ISMB/ECCB: “Developing of an objective proteomic indicator of trauma severity”
• May 2021 - Talk at the Blizard Institute: “Data driven approaches to understand the hyperacute window of trauma patients”
• October 2020 - Poster at William Harvey Research Institute Annual Review - “What similarities are there between trauma, sepsis and covid?”

Publications and Reports:

• October 2022 - NeurIPS AI for Science workshop 2022
• October 2021 - Report for the Data Study group 2020
• More papers coming soon… (x2)

Other Projects

• Created a package called EnrichOmics to automate workflowes in bioinformatics. The package allows to use the EnrichR and OpenTargets APIs to get quick results without having to copy-paste lists of genes to their websites. Info can be found on PyPI. Docs can be found here(2022)

• Project on Explainable AI: Communicating high-street bakery sales predictions using counterfactual explanations (2021) This challenge aimed to help CatsAi better serve their client (a large wholesaler) to estimate bakery orders to reduce waste and under delivery. The main tasks were to predict high-street sales based on meteorological factors and apply explainability techniques to effectively communicate their outputs to the client. Research questions we wanted to answer were:

1. Which features are predictive of sales (i.e. orders placed to the warehouse)?
2. Powerful predictive models are often difficult to interpret. Can we explain which features are important to a business owner? (XAI)

• Tutorial for Apocrita (QMUL’s cluster) (2020) Developed a tutorial to help other researchers at QMUL to use array jobs on the cluster and speed up their analyses.

• Contributed to a project for Dementia Platform UK to implement models that allow clinicians to understand drivers of dementia and predict Alzheimer using a large cohort of patients.

Contacts :)

• Linkedin
• The Alan Turing Institute website
• QMUL Institute of Applied Data Science