Real-time changes improve visual perception and experience. #shapeXT was extended to allow users to move atoms in crystal structures with the mouse. The video demonstrates changing atom positions in ice (VIII) in real time, taking SG symmetry into account. #chemistry#compchem
Google DeepMind’s AI model has, for the first time, independently proposed and confirmed a new cancer treatment approach
The system, called Cell2Sentence-Scale and featuring 27 billion parameters, discovered that the drug silmitasertibmakes tumors more visible to the immune system.
The algorithm tested over 4,000 compounds and predicted that silmitasertib boosts immune response in active immune systems.
Experiments on human cells confirmed the effect: when combined with a low dose of interferon, antigen presentation increased by 50%.
This marks the first time an AI has not only analyzed data but also formulated and verified a hypothesis with clinical potential.
Researchers from DeepMind and Yale University call it a major step toward applying artificial intelligence in biomedicine — though they note the findings still require further validation.
Skala is now available to everyone!
Why are we releasing it? Because we’re not just aiming to publish a cool paper — we’re on a mission to bring DFT to chemical accuracy using deep learning. And to make real progress, we need the community’s feedback. #compchem
Microsoft Research invites organizations of all sizes to join the DFT Research Early Access Program to explore the potential of our new Skala functional and accelerate innovation across industries through faster and more accurate density functional theory. https://t.co/KKL9gHIUez
The James Webb Space Telescope detected dimethyl sulfide in its atmosphere (on Earth that comes from marine microorganisms).
K2-18b is potentially habitable super-Earth about 2.6 times the size of our planet. It is located 124 light-years from the Solar System.
@svadrb@CorporateShark However, I wonder how this could affect the L1/L2 cache hit performance if everything relies on buffer addresses. Is there anything known?
@svadrb@CorporateShark You don't need the additional handling of descriptor sets and their pools. Moreover, you don't need to update or push descriptor sets before shader invocations. Basically, you only need to obtain the buffer addresses, which then can be transferred via push constants.
@GillessenJens Toi, toi, toi an alle, die das in Zukunft aus reinem Idealismus (?) unter Selbstaufopferung noch weiter verfolgen werden. In 3 Monaten bin ich da raus...
@GillessenJens Meine Uni hat bereits einen Exzellenzcluster, aktuell wird ein Antrag auf einen weiteren Cluster ausgearbeitet und parallel eine Bewerbung als Exzellenzuni. Letztere setzt aber erst einmal zwei Cluster voraus. Derweil ist die Finanzsituation schwierig. Prioritäten, tja. 🤷🏻♂️
@GillessenJens Mich würde interessieren, wie dieser Exzellenzwettbewerb kostenmäßig in Relation zum zur Verfügung stehenden Budget steht. Es wird viel Personal und Geld allein in die Anträge gesteckt. Mitunter gibt es finanzielle Zuschüsse in Millionenhöhe von den Ländern nur für die Bewerbung.
🚨 We are pleased to present the report into the 2nd Leibniz PostDoc survey! 🚨
This has been a long-time coming but we are happy to see this report into the careers, working lives, and mental health of Leibniz PostDocs finally available.
For more information, keep reading!
@gudrun_lotze@sabinedoering@DrLutzBoehm »Permanent positions in academia and outside academia, both with a focus on research, were rated most attractive by Leibniz PostDocs, followed by permanent positions in research management and then university professorships.«
Tja, fair enough. 🤷♂️
We apologize if you were one of the ~1,300 people who tried to access the report last week when SSOAR was down.
The good news? SSOAR lives again, and the report is available for download: https://t.co/8LuhuoF08V
#IchBinHanna#WissZeitVG#AcademicTwitter#PostdocVoice