Davis McCarthy

Valentine Svensson
Negative binomial regression and inference using a pre-trained transformer
https://arxiv.org/abs/2508.04111

When Jeffrey "graduated" from our lab to go to do a PhD at Cambridge I knew big things would come - and boy was I vindicated!

Very exciting work - rigorous and insightful (exactly as I came to expect from Jeffrey) and immediately useful for a huge number of cool projects.

Awesome 🌟, well done!

Gosh the number of times I see people testing differentially expressed genes and then applying a post-hoc fold-change threshold and I have to scream into the void WE WROTE A METHOD TO TEST THE INTERVAL NULL (...15+ years ago 🫠)

people don't actually want to test point-nulls in NHST, they want to test interval nulls and they rely on the structure of the test to define the "interval null" (*waves hands vaguely*) for them, rather than explicitly defining an interval null and testing it.

📢 Announcing the MOPITAS Autumn School on Spatial Transcriptomics Data Analysis – taking place November 12–14, in Munich. Don't miss this excellent opportunity to deepen your expertise in data science and multi-OMICS analysis - especially as a PhD student in the data science community.

ABACBS 2025 Conference Adelaide, South Australia. Nov. 24-

The day is finally here! 🎉 We’re releasing the invited speaker line-up, key dates, and lots more info for ABACBS 2025.
Check it out and share widely: www.abacbs.org/abacbs2025

Registrations and abstract submissions open next week, with abstracts due in August!

Postdoctoral Researcher in Computational Genetics The research group of Oliver Stegle looks for a postdoctoral researcher to join a collaborative project with GSK with the goal to apply computational methods to investigate the effects of rare variant...

Job alert: Join us for a postdoc in AI in genetics at @EMBL Heidelberg! Great collaboration with @Adrian Cortes @GSK, aiming to develop new tools to elucidate genetic effects using population-scale cohorts and single-cell readouts. Please share!

embl.wd103.myworkdayjobs.com/de-DE/EMBL/d...

Increasingly, I think one of the most important questions people need to ask (and answer) themselves is "How much is enough?"

Applications for this position will close on Friday 30 June

So if you or anyone you know might be interested in applying, time to get your skates on!

Nice work :D

SpotSweeper: spatially aware quality control for spatial transcriptomics - Nature Methods SpotSweeper is a spatially aware method for quality control of spatially resolved transcriptomics data that corrects for spatial confounding missed by existing methods, including both local and region...

🚨 New paper published in @natmethods.nature.com!

We introduce SpotSweeper, the first spatially-aware QC methods for spatial transcriptomics.

📰 Paper : nature.com/articles/s41...
💻 Code: github.com/MicTott/Spot...
📈 Website: mictott.github.io/SpotSweeper/

🧵👇

OurDNA Symposium 2025: Partnering for impact The OurDNA Symposium brings stakeholders together for important conversations about building the foundation for equitable genomics in Australia.

Hey Australian genetics/genomics friends: the OurDNA Symposium will be in Sydney on 14 August, just before the HGSA meeting. Learn more about inclusive recruitment for genomics and get a preview of the OurDNA variant browser! events.humanitix.com/ourdna-sympo...

I would also very much like a mixed model in edgeR 😉

afaik (given I haven't been an active developer for years), it hasn't been possible to implement a mixed model in a way that's compatible with the empirical Bayes information sharing...and for edgeR eBayes is more important if there's a tradeoff

New preprint from the lab

@tyagilab.bsky.social

Yes, also open for PhD applicants too. We accept students who apply and are enrolled through the Faculty of Medicine, Dentistry and Health Sciences. Email me if you'd like to discuss further

I cannot recommend Davis’s group more highly! In addition to excellent research, he has been a great mentor!

Come for the science, stay for the March Mammal Madness!

🔍 We're looking for:
- PhD in computational field
- Strong programming skills (R/Python/Julia)
- Genomics/bioinformatics experience
- HPC expertise

💼 What we offer:
• 3-year contract
• $98-106K + generous super
• Supportive research environment

#bioinformatics #computationalbiology #postdoc

🧵3/3

Comparison of AI-integrated pathways with human-AI interaction in population mammographic screening for breast cancer - Nature Communications Successful human-AI collaboration could greatly contribute to breast cancer mammographic screening. Here, the authors use a large-scale retrospective mammography dataset to simulate and compare five p...

We're do cool stuff: single-cell genetics, spatial transcriptomics, biomedical imaging, and (dare I say it) AI (ahem, deep learning).

EG:

Developing an AI reader to detect breast cancer: www.nature.com/articles/s41...

Spatial transcriptomics in pulmonary fibrosis: www.nature.com/articles/s41...

Research Officer - Bioinformatics Job in Fitzroy, Melbourne VIC - SEEK Seeking a Postdoc to develop computational toolkits to enable large-scale studies of single-cell and spatial 'omics and statistical genetics

📢 PostDoc opportunity in our Bioinformatics & Cellular Genomics lab at SVI! 🧬

You’d join a welcoming, supportive, and brilliant team.

Why not spend a few years in Melbourne and be part of something exciting?

Apply here: www.seek.com.au/job/84737876

#ScienceCareers #PostDoc #Bioinformatics

I'm so sorry to hear this. Atul was very kind to me when I presented as a PhD student at a conference back in ~2013. That he was so impressive scientifically but so encouraging and welcoming stays with me

Delighted that the 'flagship' manuscript on our @genesandhealth.bsky.social 44k exomes (British Pakistanis & Bangladeshis) is now preprinted. Great academic-industry collaboration. Lots of new associations (mostly additive, a few recessive) and new insights into homoz knockouts & drug discovery.

People always stop me in the street to ask: "Yoav, where are the disease-associated eQLTs? We found a lot in GTEx but we can't find anymore. Do you know where they are?"

(For the record, no one has ever asked me this, but it is a really good question!)

I think we know where they are.

My message to Trump’s NIH Director? No one in America wants us to do LESS cancer research.

No one is asking Trump to make it harder to cure Alzheimer's disease.

Yet Trump is cutting all of this NOW and demanding an $18 BILLION cut to NIH next year. Not on my watch.

Standard methods are equivalent to a flashlight, looking at each gene independently. We combine signals from multiple genes, turning a floodlight onto the genome.

Excited to share my first PhD paper in the @sbmontgom.bsky.social lab with @tamigj.bsky.social (www.biorxiv.org/content/10.1...)! Standard QTL methods treat each gene independently. But what if a single variant regulates multiple nearby genes at once - what we call “allelic proxitropy”? 🧵 ⬇️

Great work and summary, Saahi! Very well done 😁

A spatial transcriptomic atlas of acute neonatal lung injury across development and disease severity A molecular understanding of lung organogenesis requires delineation of the timing and regulation of the cellular transitions that ultimately form and support a surface capable of gas exchange. While ...

I am so excited to share our most recent work & my first big project as a graduate student! We leveraged spatial transcriptomics to study the cell-cell communication involved in human lung development. Here’s the full study: www.biorxiv.org/content/10.1...
(1/17)

Super happy and proud of this one. We and others have developed tools for dimensionality reduction for many years. MOFA (linear) and other non-linear generalisations, space/time. scDORI fuses this with GRN inference. It’s a nice and useful piece of software. Here applied to GBM biology.

underrated track

My ongoing request:

If your NSF or NIH grant was terminated--whether at Harvard or elsewhere--please report it here.

NSF: grant-watch.us/submit-nsf.h...

NIH: grant-watch.us/submit-nih.h...

Our trackers are actively used in lawsuits and are often the only record that terminations ever occurred.