Choosing your track - spectra, separations, omics, imaging, cytometry
Lesson 1 of 6 · Your instrument, end to end
1 · Learn the move · Flipped interaction
This area works differently: lessons 2 through 6 are five parallel tracks, not a sequence. Spectra and separations. Sequencing and variants. Expression and single cell. Imaging. Cytometry. You complete ONE - the one whose files actually land on your desktop - and you run it on your own data. The other four will still be here if your work changes. Every track teaches the same three moves on different formats: convert the vendor file to something open, shrink it to a reasoning slice with a control inside, and script anything you do twice. Choosing is itself the first exercise: flip the interaction and let the model interview you about your instrument, your file extensions, and your actual question, one question at a time, until the track picks itself.
Interview me to choose my pipeline track. One question at a time, wait for each answer: (1) what instrument produces the data I analyze most, (2) what file extensions land in my folder after a run, (3) what I currently do with those files, (4) what question the analysis is supposed to answer. Then recommend ONE track from: spectra/separations, sequencing/variants, expression/single-cell, imaging, cytometry - with the file formats that drove the choice. If my answers span two tracks, say which comes first and why, do not merge them.
2 · Your turn. You write the prompt
You run a shared-resource lab and touch three instruments in a bad week, so "pick your track" reads like a trick question. Let the model interview you about where your analysis hours actually go, and land on the one track to complete first - not a custom blend of all five.
Remember: the AI sees only your prompt, not this page. If the situation isn't in your prompt, it doesn't exist.
Optional. These shape the output when you run your prompt below, not your score.