London, England, United Kingdom
I am a final year PhD candidate in biomedical data science at Imperial College London with special emphasis on metabolomics. My research focuses on the integration of multimodal mass spectrometry data in metabolomics including multi-assay LC-MS data, and mass spectrometry imaging data. My general research interests lie in the utilisation of machine learning and AI methods in clinical research and precision medicine. As part of my PhD research, I created and published MAMSI (Multi-Assay Mass Spectrometry Integration), an open-source Python package on PyPI, and recently authored a first-author paper in Analytical Chemistry detailing this methodology.
In my role as a graduate teaching assistant, I am tasked with leading both supervised and unsupervised multivariate analysis tutorial within the Hands-on Data Analysis for Metabolic Profiling short-course of the Imperial International Phenome Training Centre. I am also providing learning support to BSc Biomedical Science students enrolled in the Biomedicine Data Science module. Further, I am also supporting MSc Transitional Neuroscience students within their core Introduction to Computational Methods for the Brain Sciences module. Supporting both modules entails assisting students with coding, debugging, and interpreting results of machine learning and statistical models. In addition, I have been involved in marking assessments for these modules.
As a Research Software Champion, I was tasked with supporting my fellow PhD students and colleagues in the department of Metabolism, Digestion and Reproduction with applying good coding practices in their research. I ran one-to-one coding clinics and organised a workshop focusing on version control with Git and GitHub.
During my internship, I worked for the Platform Data Science team on modelling bioactivity and mass spectrometry data, as well as developing data quality metrics. To achieve this, I applied statistical approaches such as generalised linear models and hypothesis testing, alongside supervised machine learning methods including Random Forests and support vector machines.
I was a contractor to Pfizer UK during my summer project placement at the London School of Hygiene & Tropical Medicine, University of London (LSHTM). My work included development of rule based and machine learning-based models for natural language processing that supports systematic literature review. This models were used to support a rapid literature review on 'How Pharmaceutical, Biotechnology and Medical Technology Companies use the Clinical Practice Research Datalink' that is currently in the process of dissemination.
Public Health and Social Measures (PHSM) was a joint project between the World Health Organization (WHO) and the London School of Hygiene and Tropical Medicine, University of London (LSHTM). The aim of this project was to collect non-medical interventional response to the outgoing COVID-19 pandemic. I worked on voluntary basis as programmer and data scientist. I have developed an automated input validation tool that has been used to validate volunteers data input to the PHSM database (https://github.com/kopeckylukas/PHSM_Target_Filter). Further, I was involved in data data analysis producing weekly data updates (https://github.com/lshtm-gis/WHO-PHSM).
As a part of my Data Challenge module at the London School of Hygiene and Tropical Medicine, I conducted research on how cancer waiting times within NHS England were affected by the COVID-19 pandemic within a group of four. This involved data extraction from a large unstructured dataset, visualisations using ggplot2 and using statistical modelling for predictions. The findings of the research were presented to the client at the end of the project.