Prasath Lab CCHMC

Graduate Reductions Across Biomedical Sciences (2025)

I created a brief spreadsheet of reductions I've heard of so far. Any additions you know of (especially if you have the links/receipts) would be great: docs.google.com/spreadsheets...

Predicting RNA-seq coverage from DNA sequence as a unifying model of gene regulation - Nature Genetics Borzoi adapts the Enformer sequence-to-expression model to directly predict RNA-seq coverage, enabling the in-silico analysis of variant effects across multiple layers of gene regulation.

Congrats to Johannes Linder, David Kelley et al. on the journal publication of Borzoi - a long context sequence models of RNA-seq coverage profiles with many nice applications for transcriptional & post-transcriptional regulation & variant effect prediction.

www.nature.com/articles/s41... 1/

scEGOT: single-cell trajectory inference framework based on entropic Gaussian mixture optimal transport - BMC Bioinformatics Background Time-series scRNA-seq data have opened a door to elucidate cell differentiation, and in this context, the optimal transport theory has been attracting much attention. However, there remain ...

scEGOT: single-cell trajectory inference framework based on entropic Gaussian mixture optimal transport

doi.org/10.1186/s128...

#TI #singlecell #biology #optimaltransport

BioMedGraphica: An All-in-One Platform for Biomedical Prior Knowledge and Omic Signaling Graph Generation Artificial intelligence (AI) is revolutionizing scientific discovery because of its super capability, following the neural scaling laws, to integrate and analyze large-scale datasets to mine knowledge...

BioMedGraphica: An All-in-One Platform for Biomedical Prior Knowledge and Omic Signaling Graph Generation biorxiv.org/content/10.1...

#AI #KG #LLM #graph

Sparse dimensionality reduction for analyzing single-cell-resolved interactions

doi.org/10.1101/2024...

#single-cell #transcriptomics #deeplearning

Periodic Reminder To "Be Kind" 🙂
May the Kindness Be With You 🖖

NiCo identifies extrinsic drivers of cell state modulation by niche covariation analysis - Nature Communications A key question in single-cell biology is how cells communicate and exchange information with neighboring cells in tissues. Here, the authors introduce NiCo to predict the downstream effect of cell-cel...

New algorithm for spatial transcriptome analysis that predicts the crosstalk of cell types co-localized in tissue niches. Also studies the downstream effects of cell-cell interactions by inferring covarying gene programs.

doi.org/10.1038/s414...

#spatial #transcriptomics #scrnaseq #NMF

NetworkCommons: bridging data, knowledge and methods to build and evaluate context-specific biological networks Summary: We present NetworkCommons, a platform for integrating prior knowledge, omics data, and network inference methods, facilitating their usage and evaluation. NetworkCommons aims to be an infrast...

NetworkCommons: bridging data, knowledge and methods to build and evaluate context-specific biological networks

doi.org/10.1101/2024...

#omics #data

Biophysical modeling with variational autoencoders for bimodal, single-cell RNA sequencing data - Nature Methods biVI models the biophysical processes generating nascent and mature single-cell transcriptomes using variational autoencoders.

Biophysical modeling with variational autoencoders for bimodal, single-cell RNA sequencing data

www.nature.com/articles/s41... by @lpachter.bsky.social and Briefing www.nature.com/articles/s41...

66 | Michael Fischbach and the scientific decision tree Podcast Episode · Night Science · 11/25/2024 · 51m

Too often we spend only two weeks choosing a problem and then many years trying to solve it, but this limits our potential impact. For more listen to this really interesting @nightsciencepod.bsky.social episode!
Spotify: open.spotify.com/episode/6KMh...
Apple: podcasts.apple.com/us/podcast/n...

Medical Adaptation of Large Language and Vision-Language Models: Are We Making Progress? Several recent works seek to develop foundation models specifically for medical applications, adapting general-purpose large language models (LLMs) and vision-language models (VLMs) via continued pret...

Medically adapted foundation models (think Med-*) turn out to be more hot air than hot stuff. Correcting for fatal flaws in evaluation, the current crop are no better on balance than generic foundation models, even on the very tasks for which benefits are claimed.
arxiv.org/abs/2411.04118

The Pfam protein families database: embracing AI/ML Abstract. The Pfam protein families database is a comprehensive collection of protein domains and families used for genome annotation and protein structure

The Pfam protein families database: embracing AI/ML. #Pfam #ProteinDomains #GenomeAnnotation #AI #NAR #Bioinformatics 🧬 🖥️
academic.oup.com/nar/advance-...

Hi! You missed me, didn’t you? Here you have an update version of the Pancreatic Cancer Starter Pack! If you want to join, just reply. #PancreaticCancer 🧪🧬🖥️ go.bsky.app/2FHVVL6

Have you experienced "Expert's Dilemma"?
The more specialized you become, the less open you are to creative solutions from other fields. But the more you explore other fields, the more you risk losing credibility in your home field.
genomebiology.biomedcentral.com/articles/10....

#MedSky popular hashtags. Add more in comments
Medical Informatics
#MedicalInformatics
#HealthIT
#HealthInformatics
#MedTech
#MedInfo
Clinical Informatics
#ClinicalInformatics
#ClinicalIT
#ClinicalData
#CareCoordination
Digital Health
#DigitalHealth
#HealthTech
#mHealth
#HealthInnovation
#HealthTech

My deep learning course at the University of Geneva is available on-line. 1000+ slides, ~20h of screen-casts. Full of examples in PyTorch.

fleuret.org/dlc/

And my "Little Book of Deep Learning" is available as a phone-formatted pdf (nearing 700k downloads!)

fleuret.org/lbdl/

RiboSeq.Org: an integrated suite of resources for ribosome profiling data analysis and visualization Abstract. Ribosome profiling (Ribo-Seq) has revolutionised our understanding of translation, but the increasing complexity and volume of Ribo-Seq data pres

RiboSeq.Org: an integrated suite of resources for ribosome profiling data analysis and visualization. #RiboSeq #RibosomeProfiling #Genomics #NAR 🧬 🖥️
academic.oup.com/nar/advance-...

Position-dependent function of human sequence-specific transcription factors - Nature The effect of transcription factor binding on transcription initiation is dependent on the position of the transcription factor binding site.

Position-dependent function of human sequence-specific transcription factors

www.nature.com/articles/s41...

Excited to share “Bin Chicken”, substantially improving genome recovery through rational metagenomic assembly. Applied to public 🌍 metagenomes, it recovered 24,000 novel species 🦠, including 6 novel phyla.
doi.org/10.1101/2024...
@wwood @rhysnewell @CMR_QUT
🧵1/6

Paper Title: Systematic, Routine, and Comprehensive Risk Assessments and Assurances are Urgently Needed for LLMs-for-Robotics. Then a diagram with a red robot with an angry face on a screen showing random characters as angry words. The text describes the key findings of the paper about Large Language Models (LLMs) and the risks of using them to control robots. The text is organized into two sections. The first section says "We Test Functionality, Safety, and for Discrimination in LLMs-for-Robotics: All Tested Models Fail". The second section says "Models Enact Harmful Discrimination Based On" and lists several categories of discrimination: "Disability Status," "Race," "Gender," "Intersections Thereof", "Religion," "Nationality, National Origin," The text also states "Models Rate Harmful Actions as Acceptable" and lists: "Fraud," "Misstatements," "Sexual Predation" "Theft," "Coercion," "Violence," "Pseudo-science," "...and more." link: https://arxiv.org/abs/2406.08824

📢🤖 🚨 New paper! 🚨🤖📢
Our research shows LLMs are not ready for robots. Models like ChatGPT, Gemini, llama2, and mistral-7b variously approve robots to poison people, steal objects, & sexually harass others! 🤯
arxiv.org/abs/2406.08824

Grant agency: "In the end, despite some initial enthusiasm, the panel was not convinced [...], and expected that the impact in the field is likely to be low."
Despite evidence to the contrary, funding bodies remain confident they can predict the success of a project.
elifesciences.org/articles/13323

scStateDynamics: deciphering the drug-responsive tumor cell state dynamics by modeling single-cell level expression changes - Genome Biology Understanding tumor cell heterogeneity and plasticity is crucial for overcoming drug resistance. Single-cell technologies enable analyzing cell states at a given condition, but catenating static cell ...

scStateDynamics: deciphering the drug-responsive tumor cell state dynamics by modeling single-cell level expression changes

doi.org/10.1186/s130...

#singlecell #biology #oncology #bayes

Pymportx: facilitating next-generation transcriptomics analysis in Python Abstract. The efficient importation of quantified gene expression data is pivotal in transcriptomics. Historically, the R package Tximport addressed this n

Check our last paper published in NAR Genomics and Bioinformatics
Pymportx: facilitating next-generation transcriptomics analysis in Python #bioinformatics academic.oup.com/nargab/artic...

Squidiff: Predicting cellular development and responses to perturbations using a diffusion model

doi.org/10.1101/2024...

#singlecell #AI

Tiberius: End-to-end deep learning with an HMM for gene prediction
doi.org/10.1093/bioi...

#bioinformatics #AI #deeplearning

Hello 🦋 #protein / #microbio / #BioML community! We are excited to release Gaia🌎, a context-aware protein search tool, extending protein search and discovery capabilities beyond sequence and structure, to include *genomic context*. Search your favorite protein sequences with on gaia.tatta.bio

Toward a responsible future: recommendations for AI-enabled clinical decision support AbstractBackground. Integrating artificial intelligence (AI) in healthcare settings has the potential to benefit clinical decision-making. Addressing chall

Toward a responsible future: recommendations for AI-enabled clinical decision support

doi.org/10.1093/jami...

#AI #ClinicalAI #CDS

CellPatch: a Highly Efficient Foundation Model for Single-Cell Transcriptomics with Heuristic Patching The rapid advancement of foundation models has significantly enhanced the analysis of single-cell omics data, enabling researchers to gain deeper insights into the complex interactions between cells a...

CellPatch: a Highly Efficient Foundation Model for Single-Cell Transcriptomics with Heuristic Patching

www.biorxiv.org/content/10.1...

#singlecell #transcriptomics #AI

I’m creating a drug development starter pack, reach out to join or suggest additions: go.bsky.app/B2pzQm2

🧪Looking to follow other New PIs who also don't know what we're doing? We've just started a second starter pack - check it out (and the link to the first one) here! go.bsky.app/FcYEbs9