AI & Protein Structure Prediction

#DidYouKnow that #artificialintelligence can help us better understand the function of #proteins?

🦠 Proteins are essential for sustaining life and oversee important biological processes such as cell structural support and immune protection. Some common examples are Actin, keratin, insulin, and amylase.  [1]

🧬 Each protein's biological function is defined by the protein's three-dimensional structure, which is dependent on the amino acid sequence. The prediction of the protein structure aims to determine the spatial location of every atom in a protein molecule. [2]

💻 Computational approaches of the last decade have allowed us to model many reliable protein structures and generate structure models for structure-based protein function annotation, mutation analysis, ligand screening, and drug discovery.

🔍 Many deep learning methods, such as AlphaFold, D-QUARK, and D-I-TASSER, have shown an outstanding performance in the prediction of protein structure. Specifically, in the CASP14, AlphaFold2 developed by DeepMind, showed more accurate predictions than its competitors [3].

🧠 The #AlphaFold neural network architecture and training are based on protein structure's evolutionary, physical, and geometric constraints. This network can predict certain proteins' atoms' 3D coordinates using the primary amino acid sequence as inputs [4]. 

🔬 AlphaFold has partnered with the Drugs for Neglected Diseases Initiative (DNDI) to find cures for Leishmaniasis and Chagas disease. They have also supported World Neglected Tropical Disease Day to study diseases like Leprosy and Schistosomiasis.

🔍 AlphaFold is #opensource, and can be accessed on

[1] Agnihotry, S., Pathak, R. K., Singh, D. B., Tiwari, A., & Hussain, I. (2022). Protein structure prediction. Bioinformatics, 177–188.

[2] Leach, A. R., & Thomas, P. J. (2017). Protein Structure Prediction and Homolog Modeling. Comprehensive Medicinal Chemistry III, 3–8, 120–144.

[3] Pearce, R., & Zhang, Y. (2021). Toward the solution of the protein structure prediction problem. Journal of Biological Chemistry, 297(1), 100870.

[4] Jumper, J., Evans, R., Pritzel, A., Green, et al (2021). Highly accurate protein structure prediction with AlphaFold. Nature 2021 596:7873, 596(7873), 583–589.

#deeplearning #bioinformatics  #didyouknow #ai #womenintech #alphafold #womeninai #womeninrobotics #generationequality

Contributing Editor: Women in AI & Robotics community member Astrid Carolina Padilla Arrieta

30 views0 comments

Recent Posts

See All