John Lock

Our #AI #research @NatureComms will help you use concepts from #multiomics data integration to fuse 4-channel imaging data into unlimitedly multiplexed image datasets to interrogate complex cell biology, like #EMT. rdcu.be/emtBc @UNSW @UNSWMedicine @vafaee_fatemeh

Excited to share Cancer Systems Microscopy lab @UNSWMedicine multiplexed imaging (60+ markers/cell) for #EMT research & #CTC phenotyping from #liquidbiopsy, + extensible immunofluorescence (ExIF) for ♾️ computational multiplexing via routine 4-plex IF. tinyurl.com/36x7eu6w

I'm developing a new onboarding, orientation & support pack for postdocs & professional staff @UNSWMedicine School of Biomedical Sciences. What's the #1 thing you wish someone had told you when you arrived at your institution, to help you connect, contribute & launch your career?

In academia you sometimes have to work 7 days a week, but the freedom to choose which 7 days is unparalleled

Interested in applying Prophet for your own functional assay? Check out the manuscript biorxiv.org/content/10.110… or grab the model itself: github.com/theislab/proph… Really excited about this new predictive approach for phenotype & drug discovery, check it out!

Its flexible encoding allows Prophet to predict genetic perturbations, small molecules & treatment combinations within one model. It scales w data, improving as it learns. Interestingly, learning across phenotypes boosts results, like using IC50 or RNA data to infer morphology.

By leveraging transfer learning across various phenotypes, Prophet allows in silico prediction of unseen perturbations in unseen biological contexts to predict which experiments are more likely to succeed - reducing the number to perform by up to 60x in the IC_50 example below.

Thrilled to share our „PRedictor Of PHEnoTypes“ model Prophet! Led by @Alejandro__TL & @_yji_, Prophet is a transformer-based model that predicts outcomes for unseen experiments. It aims to understands biology by learning across assays and phenotypes over 4.7M+ experiments.

I'm excited to share the next version of patchwork with the world, available on CRAN now. First class support for gt tables and even more freedom with `free()` tidyverse.org/blog/2024/09/p…

REVIEW @NatImmunol @parkerlabPMCC The complexity of immune evasion mechanisms throughout the metastatic cascade nature.com/articles/s4159…

This important review from the @parkerlabPMCC has been published in @NatRevImmunol highlighting multiple mechanisms cancers use to evade immune detection during metastatic spread & considerations to target aggressive cancers @Nicole160368 @thomasbchadwick go.nature.com/4gq0URQ

＼無料ウェビナー開催📣／ ニューサウスウェールズ大学のJohn Lock先生(@locusJ ) にご講演いただきます！顕微鏡を活用した精密医療（Precision Medicine）の最新情報をお届けします ＊本ウェビナーは英語講演です 日時：10/15 14時~ 参加方法：Zoom ▽お申込みはこちら events.zoom.us/ev/AilVvwFmO8m…

Let me introduce: InstanSeg 🦠🔬💻👩‍🔬 This *would* have been a short thread about Thibaut Goldsborough’s PhD work… but he solved too many problems. Now it's a long thread about 2 preprints, a whole new approach to cell segmentation & #opensource software to make it easy to use

Out at @naturemethods just now! nature.com/articles/s4159… We present #SN2N, a flagship level unsupervised denoising solution. It’s fully competitive with supervised learning without needing large training-set and clean ground-truth, using 1 noisy frame for training. (1/n)

For creatively thinking about & tackling really #complexproblems, which of the following do you think achieves the best process & outcomes? Why? What is required in each situation to optimise outcomes?

Mandy, always ahead of me in promoting me! I’m proud of what we achieved. It’s been a special journey! #dreamteam #phdchat

In #cellbiology, we're used to the idea of information flowing by A binding to B. But with finite numbers of A, the AB interaction reduces A's other interactions. How well do we map such negative information effects & their network propagation using epistatic perturbations?