Infectious Disease and Public Health Researcher exploring computational biology and mathematical modeling.
What can we trust in a measurement, what can a model explain, and what should we measure next?
My background spans chemistry, molecular laboratory science, epidemiology, and biostatistics. This portfolio connects those interests through public-data analyses, biological models, and tools for tracing evidence. The common thread is making the reasoning behind a result inspectable.
Three studies introduce the main question. They use different biological systems; this is a reading path, not a single analysis pipeline.
| Study | Question | What to look for |
|---|---|---|
| 1. Enzyme kinetics | Can different explanations fit the same measurements? | How the measured concentration range changes parameter uncertainty. |
| 2. GFP policy comparison | Does a more elaborate selection rule actually help? | A recorded expectation, a matched comparison, and an unexpected result. |
| 3. MAPK dynamics | Which observation would distinguish competing explanations? | A simulated cycle of choosing a measurement and updating model hypotheses. |
The green fluorescent protein (GFP) comparison tests ways to select existing brightness measurements under a fixed budget. Across four seeds, diversity-only selection had the lowest mean error on more-distant variants; uncertainty-only selection had the lowest mean error on local variants. The combined policy did not support the prediction that it would help most at distance. These are exploratory results on reused public outcomes.
The next planned analysis asks whether the combined policy's shortlist excludes useful candidates. After that, the direction is to test when measurement selection fails in the mitogen-activated protein kinase (MAPK) simulation. The roadmap separates these proposals from completed work.
- A public-health connection: Surveillance data quality examines denominators and reporting windows before interpreting a trend.
- A research workflow tool: EvidenceTrail makes extraction records and reviewer decisions inspectable. Try its browser demo with synthetic examples.
- A visual introduction: Nature's Perfume Lab explores plant aroma chemistry and protein evidence through a familiar question.
The complete project guide places all fourteen projects in this story, including sequence and expression quality, model evaluation, structure and metabolism, and evidence-review planning.
These projects use attributed public data and labeled simulations; they do not report new laboratory experiments I conducted or establish new biological findings. They were developed with AI assistance. Tests check specified software behavior; scientific interpretation remains tied to the evidence and limitations in each project.
Project guide · Roadmap · Shared standards · Glossary · Sources · Maintenance and review history