💡Thinking of using scRNA-seq to design antibody panels for spatial or single-cell profiling? Happy to launch Cytomarker, an interactive web-app and new antibody screening strategy to design and validate custom panels 1/6

🧬 Join our lab in Toronto, Canada 🇨🇦! Exciting postdoc opening at the intersection of machine learning and cancer genomics - developing next-gen models to predict genetic interactions with in-the-loop validation. Apply ➡️ or DM for more details!

RT @fabian_theis: DeepSeek isn’t just fun to use (thanks @ollama) - it’s also shaking up science, since it lets us see inside. More @Nature…

RT @sgqcheng: .@kieranrcampbell @hartjackson @SinaiHealth @MoGen_Grad @UofT @UofTCompSci @OICR_news present STARLING for cell type inferenc…

When asked about knockout vs knockdown, Google AI wonderfully bet-hedges on whether you're interested in genetics or boxing

RT @razoralign: STARLING: Segmentation aware probabilistic phenotyping of single-cell spatial protein expression data

Software available here: Great collaboration with @hartjackson with support from @SinaiHealth @OICR_news @UofTCompSci

STARLING explicitly models the consequences of segmentation errors, scalably inferring both denoised cell clusters and per-cell segmentation error probabilities, all without requiring prior knowledge of cell types or abundances (2/4)

How can we quantify single cell expression from multiplexed imaging datasets given segmentation errors? To help tackle this, we introduce evaluation metrics, a gold-standard dataset, and STARLING—a new probabilistic machine learning method (1/4)

@clintomics @MedicineUVA Congratulations!

@cwcyau I always thought more John Rebus but could see Cummings

The killer AI use case nobody is talking about (and it actually works!)

Anyone have a good reference for a paper looking at the limits of expression deconvolution of either Visium or even bulk RNA-seq wrt cell type granularity? (i.e. how granular can the cell type be and deconvolution still work)

RT @Aiims1742: The 3rd edition of this comprehensive review on the genetics and biology of #PancreaticCancer is now published in @GenesDev…

@slavov_n We teach this paper in class as an example of how "don't change more than one thing at once" is just as important in computational biology

RT @jasonhartford: This has been a really fun project to work on. Experimental design in large screening experiments looks a lot like activ…

RT @DrFranVazquez: Dear #DepMap portal users, we need your support for our NIH grant application funding the DepMap portal. If you like th…

@vitaliikl @PyTorch Guesses: - The pytorch implementation has log_input=True by default meaning input gets exponentiated, so log on LHS of original eq. is wrong - The implementation uses Stirling's approx. for the factorial -> loss can be evaluated for real numbers

@RiyueSunnyBao @NCIEytanRuppin Are these open to be joined or are the recordings available anywhere?