Fabian Schmich

machine learning in pharma @roche and @genentech | PhD from @eth | previously @icr_london, @donnellycentre and @tu_muenchen

[Nikita Makarov @nikitamakaro]

DT-GPT: showing that LLMs can forecast patient trajectories (1/10)

Now in @npjDigitalMed https://t.co/WpBPuTkwxC
Also in @EricTopol’s Doctor Penguin!

Big thanks to @MariaBordukova, @RaulREsteban Papichaya Quengdaeng Daniel Garger @fschmich @MendenMichael @HelmholtzMunich @Roche

[MUDS @MunichDS]

Excellent result from one of the MUDS-Roche collaborations! Congrats 🎉 to @Ali_S_Boushehri, @MarrCarsten, Fabian Schmich @Roche and colleagues

[Fabian Schmich @fschmich]

Our perspective on #digitaltwins in Pharma is out. Great summary by @MariaBordukova and @nikitamakaro!

[Maria Bordukova @MariaBordukova]

(1/5) Generative AI-based Digital Twins will revolutionize the way we perform drug discovery and clinical trials. I will show you below the key findings of our perspective (https://t.co/EAurRFvnfM) @nikitamakaro @RaulREsteban @fschmich @MendenMichael @Roche @HelmholtzMunich

[Bill Gates @BillGates]

The world did not do enough to prepare for a pandemic. We can avoid making the same mistake with climate change by:

1. Letting science and innovation lead the way
2. Making sure solutions work for poor countries too
3. Getting started now

https://t.co/p9m1j5QuJ7

[Eric Topol @EricTopol]

No.

[Fabian Schmich @fschmich]

Free bibles, anybody? 🙏🏽

[Ryutaro Uchiyama @RyutaroUchiyama]

I'm stunned by the depth of #coronavirus information being released in #Singapore. On this website you can see every known infection case, where the person lives and works, which hospital they got admitted to, and the network topology of carriers, all laid out on a time-series

[Fabian Schmich @fschmich]

Outline for the amazing #pancancer consortium #PCAWG projects published yesterday. Lots of exciting new insights and data! 👌🏽

[Victoria Aranda @mvicaracal]

Today, @nature, @NatureGenet @NatureBiotech @NatureComms and @CommsBio are proud to publish a compendium of papers and related content from the #PanCancer project. Dive in here: https://t.co/DWEsboP3st 1/n

[Vladimir Iglovikov @viglovikov]

I am thrilled that image Augmentation library #albumentations that was born out of @kaggle competitions by @AlBuslaev @creaf @cvtalks @Dipetm and @viglovikov became a part of the @PyTorch ecosystem!

https://t.co/oGokF8ZxeG

[Nature Medicine @NatureMedicine]

"I think right now the most thoughtful people in the industry realize that human physicians combined with AI are going to be much better than human physicians alone". Read more in a News Feature from Nature Medicine: https://t.co/42rqcAMuD6

[Raul Rodriguez-Esteban @RaulREsteban]

A new opportunity in our team. If you are interested in social media listening for healthcare applications, please apply! https://t.co/rDoCZxfLDX

[Statistician Trump @StatisticianTr2]

Sleepy Journal Editors don’t buy My Study because its “underpowered”?? Are we looking at the same tables?? My N is beautiful. P=.049. Total #HOAX Null was #rejected in favor of Alternative Fact! #easy #rstats #DataScience #epitwitter #EconTwitter #Statistics

[Fabian Schmich @fschmich]

Self supervised learning for single cell feature extraction

[PLOS Comp Biol @PLOSCompBiol]

#NewResearch from Lu et al.

'Learning unsupervised feature representations for single cell microscopy images with paired cell inpainting'

https://t.co/HDxqgBfMzX

[Yarin @yaringal]

All the slides from my Bayesian Deep Learning tutorial at MLSS 2019 Moscow, including a practical in Active Learning with jupyter notebooks (practical credit: Ivan Nazarov), are now online
https://t.co/5ZmGwUnAdQ

[Fabian Schmich @fschmich]

Our new paper is out! A generative probabilistic model for data integration and gene prioritisation.

[CBG, ETH Zurich @cbg_ethz]

We present netprioR, a new model for integrating gene–gene similarity data and prior knowledge about gene functions with gene-based covariates, such as phenotypes measured in genetic perturbation screens.
R package: https://t.co/K7OqZqSPzK
Paper: https://t.co/d4ZM4CrfRg.

[Fabian Schmich @fschmich]

Google Brain‘s 2017 https://t.co/uiF7eZQBdJ

[Hilary Parker @hspter]

Thanks to @thomasp85 for making tidygraph and ggraph, took me about 5 minutes to get a graph plot from a tidy dataset 😊 #rstats https://t.co/CEFf5ZYAu9

[Fabian Schmich @fschmich]

Single cell analysis shows CD14+CD16−HLA-DRhi monocytes to be predictive for anti PD-1 immunotherapy response. https://t.co/WsHipyBk1H

[Fabian Schmich @fschmich]

@hadleywickham You know you're famous, when an international company's data science competition features you as the placeholder name. :-)

[Frank Harrell @f2harrell]

Need to find a way to kill the use of change from baseline in ||group randomized trials. Need to estimate Y1|Y0,tx https://t.co/LD1v6PDi4f