Nathan Smits

Excited to share CpGPT, a new foundation model for DNA methylation, crafted to understand the 'language' of epigenetics! 🧬🤖 What’s CpGPT? CpGPT is a transformer model pre-trained on CpGCorpus, a huge dataset of public DNA methylation info. By integrating sequence, positional, and epigenetic data, it captures complex epigenetic interactions from even a small part of the methylome. 📄 Preprint: biorxiv.org/cgi/content/sh… Why CpGPT? DNA methylation is key for gene expression and chromatin structuring, with CpG sites often varying in diseases like cancer and aging. Here’s what sets CpGPT apart: ✨ CpGPT Highlights: 🔧 Deep Pretraining: Trained on over 100K samples from 1,500+ studies, covering 1M+ CpG sites. 🗺️ Zero-Shot Skills: CpGPT imputes methylation from minimal data, mapping different Illumina arrays to a common reference. 🐭 Cross-Mammalian Reach: When fine-tuned, CpGPT accurately imputes methylation in new mammalian species. 📈 Strong Performance: Fine-tuned CpGPT ranks 2nd overall and 1st on the public leaderboard for age and mortality prediction in the Biomarkers of Aging Challenge. Special shoutout to Lucas (@ollimacsacul) for his outstanding leadership on this project! Big thanks as well to all the co-authors (@rv_sehgal, @prof_horvath, etc.) for their dedication and swift feedback throughout. We're committed to open science and are actively preparing the code. Expect the full code and model weights to be available soon—likely within the next few weeks!

@Faulkner_Lab I'd definitely add the Kangaroo Point cliffs! Great climbing smack in the middle of the city, with great (sunset) views, and conveniently close to the best bakery in Brisbane!

After years of telling visitors to Brisbane what an awesome place it is for fun* adventures, I've decided to make a list in the order I remember them of my favourite places to go for ease of reference. Here goes ... (*definition of fun may vary)

If you use a "Random number generator" aka Nanodrop to spec your DNA/RNA, some advice: 🧬Measuring DNA/RNA in H20 = highly inaccurate/variable values. 🧬Purity ratios of samples below 20ng/ul are not reliable. NEB dived deep into when it becomes limiting: neb.com/en-au/-/media/…

so glad that our endless discussions about the nitty gritty of CRISPR screens resulted in this awesome paper - congrats @gilan_omer and everyone in @TheDawsonLab nature.com/articles/s4159…

Ohio cryptic lineage update. We’ve made no progress identifying the individual, but we have learned a few things. 1/

Very nice to show you around 😀 And unexpectedly (and sadly) tricky to talk science in my native tongue !

Imprinting in placenta via nanopore. Worth a read if you're into that sort of thing! Gratitude to all involved, foremost the co-lead authors Michalea Kindlova and Hannah Byrne (no twitter afaik). Many thanks to @QLDFamilyCohort for making this possible. biorxiv.org/content/10.110…

🚨I am recruiting for a fully-funded PhD position to work at the interface of #mRNA and cell therapies. You will join a passionate team at the @AIBNatUQ and @BaseDna mRNA facility. If you are interested send me an email or DM. Please retweet! graduate-school.uq.edu.au/project/mrna-e…

So proud that my first PhD student @robinleetroskie just passed her thesis defence with flying colours! Robin-Lee was an amazing student who made huge progress in understanding human pseudogene functions. You can read part of her work here: tinyurl.com/yf5nf64j Congrats Dr!

@ael_hardy @MotherOfPhage @L_Delanghe hahaha thanks for the trip down memory lane :)

@MotherOfPhage @L_Delanghe @natenstuff1

Lab art! I see a phamiliar phace #phage

Congratulations to Alex Carleton @Carleton_Alex on a stellar Honours thesis defence! Alex did an incredible amount of work in 10 months in my lab, establishing the basis for an exciting PhD project. #ProudPI #imnotcryingyourecrying @MaterResearch @UQ_SCMB @Faulkner_Lab

Top HiFi, middle Duplex Kit14, bottom ILMN.

Great to contribute to this paper describing @_adamewing's tool Methylartist. Methylartist enables easy visualisation of modified bases from @nanopore reads. It can generate a variety of visualisations including ridge, violin, locus and chromosome plots.

Working with @nanopore modified base call data? Consider methylartist: github.com/adamewing/meth…. Appilcations Note online now at academic.oup.com/bioinformatics…. Many thanks to @SethCheetham and Michaela Kindlova.

I am excited to be starting my own group as a joint appointment between the mRNA hub @BaseDna and @AIBNatUQ. We'll be focusing on engineering next generation mRNA vaccines and therapeutics!

Is L1 retrotransposon activity in the brain specific to only some neuronal lineages? Check out our new preprint👇where we find L1 is activated by the SOX6 transcriptional program of parvalbumin (PV) GABAergic interneuron development. 1/n biorxiv.org/content/10.110…

Had a wonderful conversation with @MadeleineMHKing about various things, including: the importance of relations between Researchers and Policy makers, how to tackle imposter syndrome, and the importance of diversity. Thank you @ScienceAU for such a wonderful opportunity #SmP2022

Thank you so much for this wonderful opportunity!

Had a blast at the Lorne @GenomeConf Saw heaps of amazing presentations and had some great chats! Also very excited about this😀