Dr Cosmika Goswami
Research Associate
Strathclyde Institute of Pharmacy and Biomedical Sciences
Personal statement
Biostatistical Researcher | Computational Biologist
I am a bioinformatician with over twelve years of experience in developing computational tools for analysing clinical healthcare data and bacterial population genetics using whole genome sequencing. My expertise lies in mathematical and statistical modelling, particularly Markov Chains, and applying machine learning techniques in bioinformatics and epidemiology.
I hold a Ph.D. in Mathematical Statistics from the Indian Institute of Technology Guwahati, where I specialised in Markov Chain modelling. My postdoctoral research took me to the University of Edinburgh and the University of Glasgow before I joined the University of Strathclyde as a Research Associate.
Area of Expertise
As a Researcher at the University of Strathclyde, I specialise in bioinformatics with a particular focus on the application of machine learning methods and AI tools to analyse complex clinical and biological datasets. My work leverages advanced computational techniques to uncover insights in areas such as antipsychotics and antibiotics, aiming to improve therapeutic strategies and drug efficacy, to understand complex biological phenomena and to develop of more effective healthcare solutions.
Machine Learning in Bioinformatics:
I am proficient in applying supervised and unsupervised machine learning algorithms to bioinformatics data, enabling the identification of patterns and predictive models for clinical and biological outcomes. By using decision trees, random forests, and support vector machines (SVMs), hidden relationships can be uncovered within datasets, contributing to the optimisation of treatment regimens in antipsychotics and antibiotic research.
AI Tools for Data Analysis:
Integrating AI tools such as neural networks, deep learning, and natural language processing (NLP) large-scale datasets, ranging from clinical records to drug response data, can be processed. These tools allow to build more accurate predictive models for personalised medicine in clinical drugs like antipsychotics, as well as to enhance antimicrobial resistance studies by predicting resistance patterns and drug interactions.
Core Skills and Tools:
- Expertise in machine learning algorithms and techniques tailored to bioinformatics
- Advanced skills in AI-driven data analysis for high-dimensional biological datasets
- Experience with data pre-processing, feature engineering, and model validation for machine learning applications
- Proficient in statistical analysis and data visualisation to communicate findings clearly and effectively
Publications
- Genomic analysis of global Staphylococcus argenteus strains reveals distinct lineages with differing virulence and antibiotic resistance gene content
- Goswami Cosmika, Fox Stephen, Holden Matthew, Leanord Alistair, Evans Thomas J
- Frontiers in Microbiology Vol 12 (2021)
- https://doi.org/10.3389/fmicb.2021.795173
- Origin, maintenance and spread of antibiotic resistance genes within plasmids and chromosomes of bloodstream isolates of Escherichia coli
- Goswami Cosmika, Fox Stephen, Holden Matthew TG, Connor Martin, Leanord Alistair, Evans Thomas J
- Microbial Genomics Vol 6 (2020)
- https://doi.org/10.1099/mgen.0.000353
- A highly conserved complete accessory Escherichia coli type III secretion system 2 is widespread in bloodstream isolates of the ST69 lineage
- Fox Stephen, Goswami Cosmika, Holden Matthew, Connolly James PR, Mordue James, O’Boyle Nicky, Roe Andrew, Connor Martin, Leanord Alistair, Evans Tom J
- Scientific Reports Vol 10 (2020)
- https://doi.org/10.1038/s41598-020-61026-x
- A role for tetracycline selection in recent evolution of agriculture-associated Clostridium difficile PCR ribotype 078
- Dingle Kate E, Didelot Xavier, Quan T Phuong, Eyre David W, Stoesser Nicole, Marwick Charis A, Coia John, Brown Derek, Buchanan Sarah, Ijaz Umer Z, Goswami Cosmika, Douce Gill, Fawley Warren N, Wilcox Mark H, Peto Timothy EA, Walker A Sarah, Crook Derrick W
- mBio Vol 10 (2019)
- https://doi.org/10.1128/mBio.02790-18
- Understanding the diversity of Staphylococcus Aureus Pathogenicity Islands (SAPIs) and their virulence genes using Whole Genome Sequencing
- Goswami Cosmika, Fox Stephen, Leanord A, Evans Thomas Ronald Jeffry, Holden Matthew
- The Microbiology Society Annual Conference, 2018 (2018)
- Genetic analysis of invasive Escherichia coli in Scotland reveals determinants of healthcare-associated versus community-acquired infections
- Goswami Cosmika, Fox Stephen, Holden Matthew, Connor Martin, Leanord Alistair, Evans Thomas J
- Microbial Genomics Vol 4 (2018)
- https://doi.org/10.1099/mgen.0.000190
Research Interests
My research focuses on:
- Investigating the impact of mental health prescriptions in acute hospital care settings.
- Modelling and managing big healthcare data from NHS and Health Protection Scotland.
- Applying machine learning and network analysis for epidemiological studies.
- Analysing genomic data from pathogenic bacteria using sequencing technologies (MinION, PacBio, Illumina).
- Developing automated Python-based pipelines for large-scale genomic and healthcare data analysis.
Current Projects
Impact of High-Dose Antipsychotic Prescribing in Acute Care Settings
- Investigating the prevalence, clinical characteristics, and associated risks of high-dose antipsychotic prescribing in hospital settings across Scotland.
- Assessing the short-term and long-term clinical outcomes of high-dose versus standard-dose antipsychotic regimens using statistical and machine learning approaches.
- Supporting antipsychotic stewardship efforts by identifying factors influencing high-dose prescribing and improving evidence-based decision-making in acute care settings.
UKAR Virtual Registry (Scotland)
- Utilising record linkage of administrative health datasets to assess the real-world use, effectiveness, and safety of recently licensed antimicrobial agents (RLAs) in Scottish hospitals.
- Investigating patient characteristics, treatment regimens, microbiological investigations, and clinical outcomes associated with RLAs.
- Supporting antimicrobial stewardship efforts by identifying gaps in current prescribing practices and contributing to more judicious use of antimicrobials.
Previous Projects
Scottish Healthcare Associated Infection Prevention Institute (SHAIPI)
- Large-scale genomic epidemiology study on pathogenic bacteria.
- Development of computational pipelines for whole genome sequencing analysis.
- Collaborative research involving multiple Scottish universities and NHS Scotland.
Molecular Epidemiology of Clostridium difficile in Scotland
- Developed novel research methods combining genomic analysis with health informatics.
- Analysed bacterial population structures and antimicrobial resistance patterns.
- Project funded by SIRN (Scottish Infection Research Network).
Genomic Selection in Sitka Spruce Trees
- Investigated genetic markers for tree breeding strategies.
- Applied statistical modelling for predicting genomic evaluation accuracy.
- Project funded by the European Commission FP7.
Contact
Dr
Cosmika
Goswami
Research Associate
Strathclyde Institute of Pharmacy and Biomedical Sciences
Email: cosmika.goswami@strath.ac.uk
Tel: 01415482478