My TEDx talk on the importance of building interpretable, open box machine learning models and using domain-informed detective work to accelerate discovery in genomics, biology, and medicine: https://t.co/3hhGRMqPUs

Our neighborhood nonnegative matrix factorization preprint just dropped! Try it out for matrix decomposition and hard clustering for your spatial transcriptomic data! Blue-hued tweetorial:

We are extending the deadline for Abstract submission to the 22nd!

Registration is free! Please join us next month: https://t.co/ah0Cpd6bPe This in-person workshop is co-organized with @BeEngelhardt, @yashasannadani, @syrineblk and sponsored by @CIFAR_News.

We have a track of lightning talks for submitted abstracts -- please encourage students and postdocs to submit their best work! Topics are broadly related to experimental design & AI for science, and are outlined in Call For Abstracts (deadline Mar. 15):

We have an exciting lineup of speakers including David Baker, @jenniferchayes, @Ashia__Wilson, @StefanoErmon, @jlistgarten, @airstreets, @arkrause, @brianltrippe, @yisongyue, @jure and @kchonyc

We’re thrilled to announce the workshop “Experimental Design: AI for Science” happening at Stanford University on 3rd and 4th April 2025. The workshop focuses on theory and methodology of AI-based experimental design and its application to science.

My NIH Study Section (MABS) was cancelled a little more than 24 hours before the start. With 35 members, & 45 h of reviews/member, that wasted 1575 hours of scientists' unpaid time. @NIH, protect scientists right now & cancel all study sections until you get clarity. @NIHDirector

Exciting update!! @bioimagearchive.bsky.social is now hosting the first publicly available Incucyte data! If you have live-cell imaging data, please consider uploading to Bioimage Archive!! Thanks to Julia Carnevale and Alex Marson for experimental data — https://t.co/A8i5P3i9VT

Check out our newest preprint on #bioxiv. We developed Patches, a tool for extracting joint and condition specific signatures from scRNA-seq data in complex experimental designs like wound healing: https://t.co/mVNlDbfevY GitHub:

Feedback welcome! And join us over in the Good Place: @thebeehive.bsky.social Try out Caliban and Occident on your own Incucyte data! More phenotypes and analyses added regularly. https://t.co/p9Rx8oYrAT https://t.co/7uP0vrxXqh

Please play with these data! There is a lot more signal there. Thank you to @bioimagearchive.bsky.social for hosting these Incucyte image data -- this is a new thing for them, and they have been so kind in working through the details of submission (link coming soon!)! 🎉

With five new collaborations in the works, and a paper characterizing the differences using explainable AI already accepted as an oral presentation at #PSB2025 (lead by high school senior Marcus Blennemann), look for future work in this space!

In sum, compared to the SH KO control condition, TCR T cells with the RASA2 KO have a longer dwell time and cripple cancer cells more effectively this way, whereas TCR T cells with the CUL5 KO proliferated more frequently upon activation, adding more T cells to the fight.

With a Markov model, we deconvolved when, in frame t-1, there is 1 cancer cell and 1 T cell in a window, and in frame t there is 1 cancer cell and 2 T cells. We quantified how often this doubling of T cells attacking a cancer cell was due to proliferation or due to recruitment.

Most thrilling is that we can identify active T cells based on relative cell size and morphology, and watch T cells activate (differentially based on condition) after interacting with cancer cells.

Even more exciting, the speed of cancer cells decreased after interactions with T cells, as did their overall size (indicating stress).

While the # of T cell--cancer cell interactions increased similarly, their effects were modulated by the CRISPR KOs. E.g., the time a T cell remained attached to a cancer cell (as estimated by a negative binomial and Markov model separately) was highest in RASA2 KO T cells.

Cancer cell and T cell morphology changes dramatically depending on state. These changes are visible in the brightfield imaging – active interacting T cells are larger and change to less circular shapes. Cancer cell begin to aggregate together when interacting with T cells.

We found that the number of T cells attached to cancer cells reduces the likelihood that the cancer cell will proliferate, with the beneficial KO T cells having greater effects on proliferation reduction.

We can study differences in cancer cell division events (lower in beneficial KO T cells) and average T cell speed (faster in beneficial KO T cells).