We've been working on mining single-cell multi-omics data using deep learning for the past 1.5 years. I'm now sharing our multi-task model, Matilda (biorxiv.org/content/10.110…), for simultaneous data simulation, dimension reduction, cell type classification, and feature selection.

How well do “mini-organs” in a dish recapitulate the human tissue? We set out to answer this question for the human eye, a super fascinating organ! I'm excited to share our preprint on bioRxiv. Check out the full story through the link! 🧵biorxiv.org/content/10.110…

🚨Job alert For those interested in #Cancer and #ComputationalAnalysis I'd highly recommend checking out this new position in the recently established #TargetDiscoveryProgram at #ProCan #CMRI ! 👉Apply here: lnkd.in/gFN_qgip

Multimodal single-cell data analyses such as simulation, visualisation, classification, & feature selection are interrelated. We introduce Matilda for doing all these analyses together. Paper: doi.org/10.1093/nar/gk…; Tool: github.com/PYangLab/Matil…

We held a reading group on Transformers (watched videos / read blog posts / studied papers by Lucas Beyer (bl16) Andrej Karpathy Chris Olah Aman Arora Jay Alammar Sasha Rush et al.), and now I _finally_ roughly understand what attention does. Here is my take on it. A summary thread. 1/n

Very excited to be at my first in-person #ISSCR2023! I will be presenting my follow-up work of the Retinal Atlas, this time generating an Embryo Atlas to look at blastoids. Please come say hello, happy to talk about my retinal and embryo work! #embryos #blastoids #computational

New research from Computational Systems Biology Lab Hao Lawrence Huang & co. University of Sydney providing a reference for future development and application of #deeplearning based feature selection methods for single-cell omics data analyses bit.ly/47xz54N

Demuxafy: improvement in droplet assignment by integrating multiple single-cell demultiplexing and doublet detection methods. url.au.m.mimecastprotect.com/s/dmyICmO5nJhp… Led by Drew Neavin in collab with the sc-eQTLGen Consortium Lude Franke Shyam Prabhakar Lab Jimmie Ye @davisjmcc Marta Melé Martin Hemberg

We are thrilled to announce the official launch of Joseph Powell's newest project, TenK10K. Thank you to the Hon. David Harris MP, Simon Giuliano, Illumina and Benjamin Kile, Garvan Institute of Medical Research for officially launching TenK10K! @cellularfutures UNSW Medicine & Health Ramaciotti Centre for Genomics

I am thrilled to have been awarded NHMRC Investigator Grant (EL1). My research will be focused on understanding the genetic regulation of autoimmune diseases at the cellular level by building a population-scale single-cell atlas of immune cells. Garvan Institute of Medical Research UNSW Medicine & Health

Garvan researchers Dr Drew Neavin and Prof Joseph Powell have been awarded a prestigious Data Insights grant from Chan Zuckerberg Initiative. The funding will support their groundbreaking 18-month project, SNAPSHOT, aimed at understanding cellular variation in health and disease.

Thrilled to share that my second project during my PhD has now been published in Nature! We introduced gsMap, a method that integrates genome-wide association studies and spatial transcriptomics for spatially resolved mapping of cells associated with human complex traits.

📢 new preprint alert: So so excited to share our analysis on the impact of common and rare variants on single-cell gene expression in blood, using WGS and scRNA-seq data from nearly 2,000 individuals and 5.4m cells as part of TenK10K phase 1 🧬medrxiv.org/content/10.110… 🧵👇 (1/n)

Want to learn about evaluating genome quality and enhance your coding skill? Sign up here! Luis Pedro Coelho is launching a live coding stream & you're all welcome to join and watch, and ask questions live. forms.gle/z4eUcXRE1e5j6S… Zoom link will be sent to registered email.

1. 🚨New preprint: tinyurl.com/tenk10k-causal. We explored causal effects of gene expression in immune cell types on complex traits and diseases by combining single-cell expression quantitative trait loci (sc-eQTL) mapping in 5M+ cells from 1,925 donors in TenK10K study and GWAS. 🧵

New preprint alert: tinyurl.com/tenk10k-multio…. Excited to share our analysis on the impact of genetic variants on single-cell chromatin accessibility in blood, using scATAC-seq and WGS from over 1,000 donors and 3.5M nuclei as part of TenK10K phase 1 🧬 🧵👇 (1/n)