Then RaptorX showed distance predicted by deep ResNet may significantly improve template-free protein folding (https://www.pnas.org/content/116/34/16856 …). That is, AlphaFold is not the first predicting distance and distance-based potential by deep ResNet.
-
-
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
RaptorX group has also implemented ResNet for distance prediction and showed that it may greatly improve protein threading inhttps://academic.oup.com/bioinformatics/article/34/13/i263/5045746 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
A nice NatureNV article. Regarding to "AlphaFold takes things a step further by changing the outputs". In fact, In Conclusion and Discussion of Ref 9 (i.e., RaptorX), the last sentence of the 7th paragraph already pointed out the next step is distance prediction.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
I read everywhere that they gave "the best structure for 25 out of 43 proteins, compared with 3 out of 43 for the next-best method". That's missleading. On deep analysis, the true factor to the second best is around 2, not 25/3. (Just being fair. I have nothing against DM!)
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
It seems like David Baker has retaken the throne by accounting for inter-residue orientation with https://www.biorxiv.org/content/10.1101/846279v1.full.pdf … with their code and model publicly available at https://github.com/gjoni/trRosetta ?
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.