Alexey Gavrikov

PhD in molecular biology, IBCh RAS.
Python, docking, deep learning, biotechnology, smFRET, and some super-resolution 🔬 with fluorogen activating proteins.

[Alexey Gavrikov @SayAlexey]

Decided to try our latest generation FAP-tag DiB3/F53L for labeling thin fibers (tired of vimentin) and this is what happened.
True live-cell (no GLOX, MEA etc) super-resolution of ensconsin (MAP7).
The acquisition took 3.5 minutes, 100 fps, 20k frames. Render drift corrected 1/

[Alexey Gavrikov @SayAlexey]

Just another #fluorescencefriday
H2B (green)
Vimentin (red)

[Octavian Ganea @octavianEganea]

New paper! Fast blind structural drug binding using geometry&deep learning.

Understanding how a small drug-like molecule attaches to a target protein is a core problem in drug discovery. By interacting with specific surface areas, drugs can change proteins' functions. 1/

[Greg Brockman @gdb]

You can now use our best models to produce embeddings (machine-readable representations) of text or code for fast & accurate search, clustering, topic modeling, and classification: https://t.co/OK8SVZTw29

[Karen Sarkisyan @k_sarkisyan]

Glowing plants are here! ✨

Bioluminescent petunia is the first product of @light_bio: enchanting everyone who sees it, glows from seeds to flowers, now being grown for commercial sales.

Currently fundrasing – to see these plants in person, DM me or [email protected]

[dana_peer @dana_peer]

(1) Finally out, after one of the most painful and long reviews (with nitpicky reviews, see my last comment) is a method that I am extremely proud of and excited about. One of the best in my career: https://t.co/H31ZezbA1J

[Kevin K. Yang 楊凱筌 @KevinKaichuang]

- Comparison of 4 homologous GFP fitness landscapes measured via mutational scanning
- Use ML to design GFPs with 48 mutations from wild type

https://t.co/C6nZSTcHgW

Louisa Gonzalez Somermeyer, Aubin Fleiss, @fkondras
@ktptntsv @a_s_mishin @krnsrksn

[Nature Methods @naturemethods]

Are you curious about using organic dyes in your microscopy experiments but not sure where to start? This Perspective from @rhodamine110 and @jonathangrimm introduces readers to the world of dyes and their pros and cons for biological applications. https://t.co/3Vm8SWB7Yz

[Fyodor Kondrashov @fkondras]

One of my best works yet: we show that the fitness landscape of GFP is heterogeneous and use it to accurately predict functional sequences with up to 20% sequence divergence. We were blown away. https://t.co/2IUQx1LTp3

[Joachim Goedhart @joachimgoedhart]

A luciferase prosubstrate and a red bioluminescent calcium indicator to image neuronal activity by Huiwang Ai and colleagues
https://t.co/CwHxFCQJ1P

[Sergey Ovchinnikov @sokrypton]

Neural networks dream of electric Proteins 🐏🐑🐏🐑🐏🐑

now with experimental validation 🥳

https://t.co/LBBgC3BcvM

[bioRxiv @biorxivpreprint]

AlphaFill: enriching the AlphaFold models with ligands and co-factors https://t.co/g5CCfzm46L #bioRxiv

[Paw @luckymarmot]

Paw · The most advanced REST & GraphQL API client for Mac https://t.co/9urIFDWKje · Our crazy🤯 2021 Black Friday offer: Retweet this and get a free Paw license (worth $49.99) 💙

[Jue Wang @jueseph]

New preprint! We came up with 2 methods to design de novo scaffold proteins to hold arbitrary functional motifs: hallucination (optimizing a seq against predictions of RoseTTAFold (RF)) and inpainting (recover masked regions of seq+struc.) 1/n https://t.co/bb0fTELr28

[RoderickTas @roderick_tas]

Happy to share that our latest work is published online in @NanoLetters. We report a Small Peptide–Protein Interaction Pair for Genetically Encoded, Fixation Compatible Peptide-PAINT.
With @Fluo_renzo @IKVoets in the @ICMStue

https://t.co/CY0sDguXAz

[Alexey Gavrikov @SayAlexey]

New fluorogen-activating tag in DiBs family! Computationally designed in Rosetta. Nice live-cell super-resolution and also a split-version. @PLOSCompBiol @MeilerLab
https://t.co/C0Sr0ENIpj

[Alexey Gavrikov @SayAlexey]

Our review about exchangeable fluorescent tags with a focus on super-resolution microscopy. Various PAINT methods, imaging by peptide-peptide interactions, fluorogen-activating proteins and more.

[Maxim Perfilov @Baragor]

Pleased to share our new review "Transient Fluorescence Labeling: Low Affinity—High Benefits" with @SayAlexey @a_s_mishin and prof. K.A. Lukyanov https://t.co/KW87UgFR7z @IJMS_MDPI @MDPIOpenAccess

[Alyona Minina @minina_alyona]

check out our preprint about SPIRO, Smart Plate Imaging Robot: https://t.co/OJjdDB04Y9
Automated time-lapse imaging is now possible inside standard plant growth cabinets.
The robot costs less than 200 Euro and can be built by a common variety biologist (we tested it!)

[fly51fly @fly51fly]

[LG] Structure-based protein design with deep learning
S Ovchinnikov, PS Huang [Harvard University & Stanford University] (2021)
https://t.co/VwIFDsZO3k
#MachineLearning #ML #AI #Bioinformatics