ShiftScope (solo)
ShiftScope compares any two single-cell populations, measures how far apart they are, finds the genes driving the difference, and ranks which genes to test next.
Core trick
Measures distance between two single-cell populations, ranks driver genes by effect size against live PubMed counts, then asks Claude for keep-or-skip.
Tags
Description
ShiftScope compares any two single-cell populations, measures how far apart they are, finds the genes driving the difference, and ranks which genes to test next. It is condition-agnostic, so the same tool handles perturbation, disease, development, or cell-type comparisons.
Its main use runs on Alex Marson's genome-scale CD4+ T-cell CRISPR screen. ShiftScope ranks hits by how strong the effect is and how little the gene has been studied, using a live PubMed count so novelty is grounded, not guessed. Claude then gives a keep-or-skip verdict. Famous regulators like VAV1 and CD45 drop; strong but overlooked genes, including chromatin-complex members with almost no papers, rise as candidates worth testing. It recovers correct biology across four domains and two real screens.
Why it matters: a screen yields thousands of hits but a lab can test only a few. These are hypothesis-generating nominations that point wet-lab effort toward high-value, overlooked biology.
aenorhabditis6/claude_science_hackathon★ 00 forksJupyter NotebookMITCreated Jul 7, 2026 · last push Jul 16, 2026
README excerpt
# ShiftScope **Compare any two groups of single cells. Measure the shift, find the genes that drive it, and let Claude tell you what to test next.** [](https://colab.research.google.com/github/aenorhabditis6/claude_science_hackathon/blob/main/quickstart.ipynb)   Built for **Built with Claude: Life Sciences** (Anthropic × Gladstone). It is condition-agnostic: control vs perturbation, disease vs healthy, young vs old, cell type vs cell type. One tool, one workflow. ## See it in 20 seconds Click the Colab badge above and **Run all**. Or, anywhere: ```python from shiftscope.quickstart import run run() # ctrl vs IFN-β monocytes -> E-distance, driver genes, UMAP, and Claude's read ``` One line loads a real dataset, measures the shift, localizes it, finds the drivers, and (with a Claude key) writes the interpretation. ## One tool, four fields The same code runs across immunology, host–pathogen disease, blood development, and cell-type identity, and recovers the correct, distinct biology each time. Nothing is hard-coded to one disease.  ## The useful part: what to test next Measuring a shift is easy. The real question is *which of hundreds of screen hits is worth a month at the bench.* On Alex Marson's genome-scale