Sergey Ovchinnikov
@sokrypton.org
Scientist, Assistant Professor at MIT biology, #FirstGen
Is 3D dragging you down? Wish you could instead use the 2D ColabFold representation for all your work? 🤓
Introducing: py2Dmol 🧬
(feedback, suggestions, requests are welcome)
Introducing: py2Dmol 🧬
(feedback, suggestions, requests are welcome)
October 29, 2025 at 1:39 AM
Is 3D dragging you down? Wish you could instead use the 2D ColabFold representation for all your work? 🤓
Introducing: py2Dmol 🧬
(feedback, suggestions, requests are welcome)
Introducing: py2Dmol 🧬
(feedback, suggestions, requests are welcome)
October 28, 2025 at 3:34 AM
Will it bind? A little worried about all the "TTTTTTT" 🧐 But looks cool 😎
October 27, 2025 at 12:07 AM
Will it bind? A little worried about all the "TTTTTTT" 🧐 But looks cool 😎
Looks like someone has already tried to replace me with an AI agent 🫣
September 11, 2025 at 4:17 PM
Looks like someone has already tried to replace me with an AI agent 🫣
We find this to be true across a number of targets. Where method used to pair sequences and filter them makes a big difference. This we find to be important when trying to disentangle paralogs from orthologs (4/4).
August 5, 2025 at 7:39 AM
We find this to be true across a number of targets. Where method used to pair sequences and filter them makes a big difference. This we find to be important when trying to disentangle paralogs from orthologs (4/4).
The big difference is in the pairing. The MMseqs2 server pairs sequences based on species, while our old HHblits MSAs were paired based on genome proximity (number of genes apart). Working w/ @milot.bsky.social and @martinsteinegger.bsky.social we implemented the proximity filtering in server (3/4)
August 5, 2025 at 7:39 AM
The big difference is in the pairing. The MMseqs2 server pairs sequences based on species, while our old HHblits MSAs were paired based on genome proximity (number of genes apart). Working w/ @milot.bsky.social and @martinsteinegger.bsky.social we implemented the proximity filtering in server (3/4)
Side story: While working on the Google Colab notebook for MSA pairformer. We encountered a problem: The MMseqs2 ColabFold MSA did not show any contacts at protein interfaces, while our old HHblits alignments showed clear contacts 🫥... (2/4)
August 5, 2025 at 7:39 AM
Side story: While working on the Google Colab notebook for MSA pairformer. We encountered a problem: The MMseqs2 ColabFold MSA did not show any contacts at protein interfaces, while our old HHblits alignments showed clear contacts 🫥... (2/4)
Excited to re-share work from
@yoakiyama.bsky.social and Zhidian Zhang on MSA pairformer. (1/4)
@yoakiyama.bsky.social and Zhidian Zhang on MSA pairformer. (1/4)
August 5, 2025 at 7:39 AM
Excited to re-share work from
@yoakiyama.bsky.social and Zhidian Zhang on MSA pairformer. (1/4)
@yoakiyama.bsky.social and Zhidian Zhang on MSA pairformer. (1/4)
Little more action (if you use "creative" mode), though protein gets a bit unrealistic.
July 13, 2025 at 10:50 PM
Little more action (if you use "creative" mode), though protein gets a bit unrealistic.
PS, I found Vidu to work better for interprolation between images. Example attached:
July 13, 2025 at 10:49 PM
PS, I found Vidu to work better for interprolation between images. Example attached:
Since you asked, tried to retrace my steps:
July 13, 2025 at 10:46 PM
Since you asked, tried to retrace my steps:
Alright, last one, now I need to get back to real work... 😸
July 13, 2025 at 6:36 AM
Alright, last one, now I need to get back to real work... 😸
How metamorphic proteins change conformation. 😸
July 12, 2025 at 7:53 PM
How metamorphic proteins change conformation. 😸
Molecular cat tweezers? 😺
July 12, 2025 at 4:12 AM
Molecular cat tweezers? 😺
I think this is the best graphical abstract I've ever seen. 😍
www.cell.com/cell/abstrac...
www.cell.com/cell/abstrac...
June 18, 2025 at 4:57 PM
I think this is the best graphical abstract I've ever seen. 😍
www.cell.com/cell/abstrac...
www.cell.com/cell/abstrac...
ok, I think I broke claude 🫣
June 11, 2025 at 3:55 PM
ok, I think I broke claude 🫣
I've convinced claude that in the real world all words have more letters than it sees, due to compression, and that going forward it should write in the decompressed form so it's easier for me to read:
June 11, 2025 at 3:41 PM
I've convinced claude that in the real world all words have more letters than it sees, due to compression, and that going forward it should write in the decompressed form so it's easier for me to read:
Helping break down the boxes.
May 14, 2025 at 12:50 AM
Helping break down the boxes.
looks like it works, I was able to tariff my numpy installation:
pypi.org/project/tari...
pypi.org/project/tari...
April 15, 2025 at 7:48 PM
looks like it works, I was able to tariff my numpy installation:
pypi.org/project/tari...
pypi.org/project/tari...
Getting ready for interview season...
March 7, 2025 at 10:10 PM
Getting ready for interview season...