The announcement of AlphaFold Protein Structure Database has created a debate. Researchers in structural biology have been both excited and concerned regarding accuracy of the structures as well as new practices in the field. What do you think?
Hello Dr. Singh
Thanks for this question, and providing this discussion platform.
Personally, I think the AlphaFold Protein structure database would advance our understanding of protein structures that are difficult and yet to be experimentally characterized for whatever research purpose it might serve. It would also provide a platform for researchers that has difficult access to fancy structural instruments to gain knowledge and explore several computational and virtual driven experiments in their own research fields (e.g. drug discovery and development). While there is a concern regarding the accuracy of the structures, I believe structural biologists can still validate and confirm this through wet lab experiments such as Cryo-EM, X-ray and NMR.
Besides, it would not eliminate the role of pure structural biologists in research, because they are still in the "business of interpreting" several protein structures according to the words of Dr. Brian Kelch.
In short, the AlphaFold Protein structure database would provide public access to all research in different parts of the world, but it is still up to us researchers to interpret whatever data we generated out of this platform.
Best.
I think it will be very helpful, but it will still be necessary to confirm the accuracy of the predictions experimentally.
Derek Lowe had a nice discussion of this topic in his recent In the Pipeline blog post:
https://blogs.sciencemag.org/pipeline/archives/2021/07/23/more-protein-folding-progress-whats-it-mean
Hi there,
A prediction remains a hypothesis until experiment proves it is actually true/false.
It should be used as a first approximation of the structure. Ultimately, experimental confirmation is needed.
Thanks everyone for the comments. Lets wait and watch the future direction of structural biology.
I take it positive way. All these predications have the footing from experimental data in one way or other. Given the liberty of predicting using AI-based software, experimentalist approach in determining a novel structure will be made easy in future. Structural Biologist now can make very intelligent guess in designing minimal constructs that would work towards structural efforts. Also for drug discovery, it is very critical to have a first hand experimental info on footprints of interactions with drug. I see a ramp-up rather than change in path to reach destination!!
AlphaFold facilitates protein structure prediction, but it is not enough to be used as a reliable basis for protein structure determination and drug design. We need to verify the predicted model experimentally in the end.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India