About Me

Welcome! My name is Margarida Rosa, but you can call me Maggie :) and my preferred pronouns are she/her. I am originally from Portugal, but I completed my Bachelors and Masters in London, and now I am in NYC pursuing a Ph.D. at Weill Cornell Medicine under the supervision of Prof. George Khelashvili.

My research interests lie at the intersection of advanced computational biophysical methods, such as machine learning and Markov state models, and their application to understanding biologically relevant systems.

Education

Weill Cornell Medicine – Cornell University, New York, NY (2021-Present)

• Ph.D., Physiology, Biophysics & Systems Biology (PBSB) program
• Mentored by Dr. George Khelashvili
• Committee Members: Harel Weinstein DSc, Olga Boudker PhD, Daniel Heller PhD

University College London, London, UK (2019-2020)

• M.Sc., Drug Discovery and Development awarded First Class-Honors (4.0 GPA)

Queen Mary University of London, London, UK (2016-2019)

• B.Sc., Biochemistry awarded First Class Honors (4.0 GPA)

Publications

• ‘Automated Collective Variable Discovery for MFSD2A Transporter from Molecular Dynamics Simulations’ Oh, Myongin†, Margarida Rosa†, Hengyi Xie, George Khelashvili. † denotes equal contribution Biophysical Journal, June 2024 Link

• ‘Dynamic Profiling of β-Coronavirus 3CL Mpro Protease Ligand-Binding Sites’ Eunice Cho, Margarida Rosa, Ruhi Anjum, Saman Mehmood, Mariya Soban, Moniza Mujtaba, Khair Bux, Syed T. Moin, Mohammad Tanweer, Sarath Dantu, Alessandro Pandini, Junqi Yin, Heng Ma, Arvind Ramanathan, Barira Islam, Antonia S. J. S. Mey, Debsindhu Bhowmik, and Shozeb Haider J. Chem. Inf. Model. 2021 Link

Selected Awards and Distinctions

• Biopysical Society Meeting Travel Award issued by Biopysical Society Meeting (Feb. 2024)
• 1st Place for Best Poster Presentation titled “Automating Collective Variable Discovery from Molecular Dynamics Simulations using Machine Learning” issued by Weill Cornell Medicine Departmental Recruitment (Jan. 2024)
• Markey Graduate School of Medical Science Fellowship, awarded to top 13% of applicants in recognition of exemplary academic and research achievements as a trainee at Weill Cornell Medicine. (2023-2024)
• 3rd Place for Best Poster Presentation titled “Human excitatory amino acid transporter 3 (hEAAT3) cation selectivity investigated by computational experiments” issued by Weill Cornell Medicine Departmental Retreat (Nov. 2022)
• Finalist at 3-Minute Thesis Competition (3MT) at Weill Cornell Medicine (Dec. 2022).
• 1st Place for Best First-Year Poster Presentation titled: “Lipid Scrambling Mechanisms of β-1 Adrenergic Receptor (β-1AR) Revealed by Computational Experiments” under Prof. George Khelashvili’s mentorship and issued at the Vincent Du Vigneaud Symposium.
• Continuous Learner Achievement issued by Roche Pharmaceutical Company (Jun. 2021)
• Honors for Master’s Thesis titled “Structural modelling of the SARS-CoV-2 RNA-dependent RNA-polymerase (RdRp) enzyme, an essential protein for viral replication” under Prof. Shozeb Haider’s mentorship and issued by University College London (Dec. 2020)
• Honors for Master’s Degree issued by University College London (Dec. 2020)
• Honors for Biochemistry Degree issued by Queen Mary University of London (Jul. 2019)

Presentations

Invited Talks

• Rosa M. “The molecular mechanism of Li+ Inhibition in MFSD2A-Mediated Lysolipid Transport”, Talk, Research in Progress Seminar Series at Weill Cornell Medicine. May 2025. New York City, NY.
• Rosa M. “The molecular mechanism of MFSD2A: a potential gateway for drug delivery to the brain” Talk, Physiology, Biophysics and System Biology (PBSB) Recruitment at Weill Cornell Medicine. February 2025. New York City, NY.
• Rosa M., Shore D., “Applications of Dimensionality Reduction Techniques” Lecture, Quantitative Biology II Course, Weill Cornell Medicine. October 2024. New York City, NY.
• Rosa M. ‘Persuading the Blood-Brain Barrier Bouncer to let you into the Brain Party’, Talk, 3-Minute Thesis (3MT) Competition at Weill Cornell Medicine. 2022. New York City, NY. Finalist
• Rosa M., Haider S. ‘Investigating SARS-CoV-2 RNA-dependent RNA polymerase (RdRp), an essential protein for RNA replication and potential drug target’. Talk, School of Pharmacy University College London. 2020. London, UK. Honors

Selected Poster Presentations

• Rosa M., Oh M., Khelashvili G. ‘Automating collective variable discovery from molecular dynamics simulations using machine learning’, Poster, Biophysical Society (BPS) Meeting. 2024. Philadelphia, PA. Travel Award Winner
• Rosa M., Oh M., Khelashvili G. ‘Automating collective variable discovery from molecular dynamics simulations using machine learning’, Poster, Physiology, Biophysics and System Biology (PBSB) Recruitment at Weill Cornell Medicine. 2024. New York City, NY. 1st Place Award for Best Poster Presentation
• Rosa M., Qiu B., Boudker O., Khelashvili G. ‘Human excitatory amino acid transporter 3 (hEAAT3) cation selectivity investigated by computational experiments’, Poster, Physiology, Biophysics and System Biology (PBSB) Department Retreat at Weill Cornell Medicine. 2022. New York City, NY. 3rd Place Award for Best Poster Presentation
• Rosa M., Khelashvili G. ‘Lipid scrambling mechanisms of β1- adrenergic receptor (β1AR) revealed by computational experiments’. Poster, 41st Vincent du Vigneaud Research Symposium at Weill Cornell Medicine. 2022. New York City, NY. 1st Place Award for Best Poster Presentation

Software Experience

During my current PhD research I developed an open-source Python-based machine learning framework using Linear Discriminant Analysis for automating collective variable (CV) discovery, reducing bias of CV selection for MD simulations. Published in Biophysical Journal, 2024. Software is publicly available in here Link. I have also designed and implemented HPC-optimized Python and Tcl/tk scripts for of large-scale MD data analysis and integrated advanced computational biophysics methods including time-independent component analysis, deep convolutional neural networks, and Markov State Models to predict MFSD2A transporter transition states and free energy landscapes.

Computational skills:

• Software Development: Machine learning (scikit-learn, xgboost, niapy), feature selection, statistical data analysis, regression, dimensionality reduction (PCA, LDA, tiCA).
• Programming & Data Analysis: Python, MATLAB, R, tcl, UNIX, bash/csh shell scripting
• Molecular dynamics (MD) & Modeling: VMD, NAMD, OpenMM, Schrödinger software suite, GROMACS, AMBER, ezCADD, Pymol, Chimera, Modeller, Autodock, LigPlot+, Coot
• High-Performance Computing: Slurm scheduler, Bash scripting, large-scale simulations.
• Certifications: Molecular Modeling in Drug Discovery (Schrödinger); Free Energy Calculations for Drug Design with FEP+ (Schrödinger); Introduction to Machine Learning (Duke University); Programming Using Python (University of Pennsylvania).

Industry Experience

I have always wondered how drugs interact with our bodies to produce the desired effect, and how technology can be used to accelerate drug development and delivery. In the pursuit of answers, I completed a 6-month internship at Roche, in Basel Switzerland, where I used unsupervised ML algorithms for drug substance development.

Contact Details

I am passionate about applying my knowledge and experience into real-life contexts, and pursuing a lifelong learning career where I can continuously master new skills and learn from others around me. If you have any ideas and want to collaborate feel free to reach me at:

Email

LinkedIn

Updated References can be provided upon request: Roche Internship Reference;
QMUL Thesis Reference;
QMUL Advisor Reference.