High costs motivate efforts to predict spatial transcriptomics from H&E images w/ deep learning. In our recent preprint, we show that noise, sparsity & resolution in ST data impact performance, highlighting the importance of training data quality: https://t.co/nJfK0qUZhx 🧵1/n

What is RNA velocity in situ? ChatGPT kept getting it wrong... So I made this video explaining how RNA velocity in situ infers gene expression dynamics by distinguishing nuclear vs. cytoplasmic expression in spatial transcriptomics data: https://t.co/IyvbNrZNBM #AcademicTwitter

When interviewing for faculty positions, it can feel like there's a "right answer" to "win". But the reality is: there’s no perfect script. Teach your ideas with clarity and enthusiasm. Trust that your genuine self will resonate more than performing what you think others want.

Looking to prep for faculty job interviews? Try my Academic Interview Simulator: Faculty Edition. About as realistic as dating sims are to real dating: https://t.co/wAdyBNjcQN (Tinkering w/ AI. Inspired by the Rising Stars workshop for post-docs I just served on. Real advice 👇)

In this blog post, I use #RStats to explore publicly available ICE arrest data. #Dataviz shows recent trend where ICE are primarily targeting/locating people without criminal records in communities. Code along and take a look for yourself: https://t.co/OzoId1DYy6 #CodeTutorial

Wrapping up AI in Molecular Biology, we’re proud to recognize three Future of Science Fund Award winners: @RyantheShark (@GladstoneInst), @reetm09 (@UCSF) & Lucy Luo (@NUFeinbergMed). Their research highlights a bright future for science! 🌟 #KSAIBio26 @drklly @JEFworks

Greetings from beautiful Sante Fe for the AI in Molecular Biology @KeystoneSymp #KSAIBio26 Excited to bring together academic and industry leaders in this emerging field to organize, shape trends, influence policy, discuss challenges If you're around, please come say hi 👋

Congratulations to @caleb_hallinan @JHUBME for leading this work in collaboration with @last_name_lucas @JHUPath 🥳🥂

While recent efforts to predict spatial transcriptomics from H&E images w/ deep learning have focused on improving modeling approaches, our results highlight that improvements in training data quality offers an orthogonal strategy to enhance performance. 7/n

Likewise, to pinpoint what imaging data quality factors may drive these performance differences, we simulated lower-resolution images by applying Gaussian blur to the H&E images to demonstrate that image resolution has a measurable impact on performance and interpretability. 6/n

Further, we demonstrate how imputation methods intended to rescue sparsity/noise boost performance when evaluated on the held-out test set but decreases performance when evaluated on an independent replicate, suggesting overfitting that limit robustness and generalizability. 5/n

To pinpoint what molecular data quality factors may drive these performance differences, we performed a series of in silico ablation experiments in which we systematically decrease molecular data to show sparsity and noise as drivers of decreased performance. 4/n

We train identical models using matched ST datasets from different technologies (Visium vs Xenium) with unique technical constraints that impact data quality. We find that the predictive performance across genes is 38% higher when trained on Xenium compared to Visium data. 3/n

Deep learning can predict gene expression from H&E, but performance varies widely, highlighting the need for further investigation into factors that impact prediction performance. Our study assesses the impact of training data quality on the predictive performance. 2/n

See the previous preprint post for more details

Further, we believe this paper will benefit from eLife's continuous post-publication public peer review. We hope this will enable the rapid dissemination of these insights + allow for constructive criticisms from industry + academic experts to be openly considered by all readers.

...we chose not to submit this as a Matters Arising since that process requires confidentiality and can be rather lengthy. Given the urgency imposed by the cost and wide usage of these ST technologies, we felt it was more responsible to share this work as quickly as possible.

I would also like to note that, given the results presented in our paper, previous results presented in the publication describing the Xenium technology and demonstrated using this gene panel (Janesick et al, Nature Communication, December 2023) are in part erroneous. However...

eLife Assessment: "This valuable study identifies and characterizes probe binding errors in a widely used commercial [ST] platform...The authors provide convincing evidence...[T]his work provides an essential quality control resource that will improve data interpretation"

Our paper identifying evidence of off-target probe binding in the 10x Genomics Xenium Breast Gene Panel is now available as a reviewed preprint at #eLife https://t.co/dQ9FqTMCdY We look forward to revising the paper to incorporate reviewer recommendations and other updates 🧵👇