Implications of AlphaFold2 for crystallographic phasing by molecular replacement

Area of Science:

• Structural Biology
• Computational Biology
• Biophysics

Background:

• The Critical Assessment of Structure Prediction (CASP) is a community-wide effort to assess the accuracy of protein structure prediction methods.
• Recent advancements in deep learning and correlated mutation analysis have led to highly accurate *in silico* protein structure models.
• Accurate protein models have the potential to impact various areas of structural biology, including crystallographic phasing.

Purpose of the Study:

• To evaluate the impact of highly accurate *in silico* protein models on crystallographic phasing methods.
• To explore the potential of using *in silico* models for molecular replacement phasing.

Main Methods:

• Analysis of protein structure models from the Critical Assessment of Structure Prediction (CASP) 14.
• Assessment of model accuracy using root-mean-square deviation (RMSD).
• Exploration of molecular replacement phasing strategies using predicted protein structures.

Main Results:

• AlphaFold2 demonstrated high accuracy in predicting protein structures at CASP14, irrespective of sequence similarity to known structures.
• The availability of accurate *in silico* models suggests a paradigm shift in structural biology.
• The study explores the feasibility of using these models for molecular replacement phasing.

Conclusions:

• Highly accurate *in silico* protein structure prediction is becoming a reality.
• The advent of reliable computational models is poised to influence and potentially simplify crystallographic phasing techniques.
• Molecular replacement phasing using *in silico* models shows promising prospects.