Julia Milena

AI&Life scientist. Recently left the irony tower to work for a unicorn 🦄. Ex:@HHMIJanelia & @ETH_en & @mpi_mr_hd & @google Intern

[Julia Milena @julia_buhmann]

And many thanks to everyone who made this possible: @SheridanArlo ,@unidesigner, @Reni3006, @uni_matrix, @herrsaalfeld, @neptunes5thmoon, @darbly, @herrsaalfeld, @gsxej, @srinituraga,, @dddavi, @janfunkey . Really happy to work with you !

[Jan Funke @janfunkey]

Automatic prediction of synaptic links in a whole Drosophila brain: Read about @julia_buhmann's method in our preprint, result of a fruit(fly)ful collaboration with the labs of @ddavi, @gsxej, @herrsaalfeld, and @darbly!

#connectomics #machinelearning
https://t.co/dsiEU7WnLA

[Recursion @RecursionPharma]

Today, we’re announcing an important step forward in modeling binding affinity alongside @MIT and a research team led by Regina Barzilay.

Boltz-2 is the first open-source co-folding model to combine protein structure and binding affinity predictions – measuring how strongly a small molecule drug will bind to a target protein.

🔹 Highlights:

▪️ Boltz-2 goes beyond AlphaFold 3 and Boltz-1 by jointly modeling complex structures and binding affinities

▪️ It’s the first deep learning model to approach the accuracy of physics-based free-energy perturbation (FEP) methods, while running 1000x faster

▪️ It’s particularly useful in hit discovery and hit-to-lead, computationally predicting binding affinities directly from molecular structure and enabling teams to rapidly prioritize which molecules to synthesize and test

Boltz-2 code, model weights, and processing pipelines will be released under an open source MIT license, making this advance accessible to every scientist. 👇

[Sven Dorkenwald @sdorkenw]

Incredibly proud and grateful to be part of the #FlyWire team that made this happen. This was an immensely collaborative project - on every level from early open data sharing, to tool building, to analysis - over many years, across many institutions. It's been a fantastic journey

[Prof. Anima Anandkumar @AnimaAnandkumar]

Our StrassenNet had shown 99.5% reduction in number of multiplications in #deeplearning and automatically discovering Strassen's algorithm. Good to see this line of work being extended https://t.co/tgxDuHqN0R

[Moritz Helmstaedter @mh_lab]

When history defines your thinking, and 80-year old events are painfully present: Why we just inaugurated a memorial for 40 killed children at our institute. A thread.

[Shoshana Zuboff @shoshanazuboff]

Friends, Musk/Twitter headlines are all variants of ‘what will Elon do?’ It’s a signal of how lost we are. We obsess over one man and his whims because we don’t yet have the democratic rule of law needed to govern our information spaces. Without law power is dangerous.🧵[1/8]

[Alex Strudwick Young @AlexTISYoung]

New educational attainment (EA) GWAS out today: https://t.co/YmgRosompg. We expand the sample size from ~1 million to ~3 million, making it one of the largest GWAS to date. We identify more loci affecting EA and increase our ability to predict EA from genetic data, and more...🧵

[Julia Milena @julia_buhmann]

@j2bryson Almost ;) women‘s day is only a holiday in Berlin, not in the rest of Germany. Maybe one day… 🤞🏼

[Reid Olsen (Seize the Means of Transduction) @Reid_dot_com]

Attention!

Are you or do you know any medicinal. chemists?

What about computational chemists?

Interested in working with @exscientiaAI?

We have positions at both our Oxford and Miami locations!

Too many links, DM me for details and please RT!

[Julia Milena @julia_buhmann]

Liebe wahlberechtigten Auslandsdeutschen, hier Briefwahl beantragen, idealerweise heute noch abschicken (natürlich per Brieftaube, digital geht nicht) https://t.co/AD4rCMJpAq

[Edward Snowden @Snowden]

Apple’s proposal to make their phones inform on and betray their owners marks the dawn of a dark future.

https://t.co/k3SxaFSJEJ

[Julia Milena @julia_buhmann]

@jmrko @mpiNeuro Congrats! Well deserved, and all the best for this next big step! 🎶

[Jörgen Kornfeld @jmrko]

Honored to open my lab at @mpineuro -- we use volume EM & computational methods to investigate how a bird learns & stores its song in the connectome, and whether we can decode it again. We are hiring :)!

[Aniruddh Galgali @argalgali_neuro]

What can trial-by-trial neural "noise"/variability inform us about recurrent dynamics and distributed population-level computations? Really happy to present a preprint of the work I did during my PhD!

https://t.co/aiu6TqpWoh

[Julia Milena @julia_buhmann]

@janfunkey @BobQubit And here is our outstanding connectomics phd hat that we made for @BobQubit. Joint work with moritz, @ssinhaleite and @lenka_cernik 🎉

[Julia Milena @julia_buhmann]

@janfunkey @BobQubit Connectomics++! Outstanding work indeed, congrats to both of you!

[Edward Snowden @Snowden]

For those wondering about @SignalApp's scaling, #WhatsApp's decision to sell out its users to @Facebook has led to what is probably the biggest digital migration to a more secure messenger we've ever seen. Hang in there while the Signal team catches up.

[Abhishek/Abhi Banerjee @abhii_mit]

First lab meeting of the year. Asked for anonymous feedback from students on mentoring style, how the lab is being run and addressed them live! Great experience, and will do this every 4 months. Intentional learning with bottom-up feedback - absolute key.

[Gregor Schuhknecht @GregorFPS]

My main PhD paper from Kevan Martin’s lab @ETH @UZH_ch now out in @nature: https://t.co/s6F7QItC20.

Combining electrophysiology and electron microscopy of the same neocortical synapses reveals multivesicular release and finally shows how synapse size predicts synaptic strength: