Tracing beta strands using StrandTwister from cryo-EM density maps at medium resolutions

Area of Science:

• Structural biology
• Biophysics
• Computational biology

Background:

• Cryo-electron microscopy (cryo-EM) enables determination of molecular structures.
• Medium-resolution cryo-EM maps (5-10 Å) allow detection of major secondary structures like alpha helices and beta sheets.
• Detecting beta strands from medium-resolution cryo-EM maps is challenging and crucial for atomic model building.

Purpose of the Study:

• To introduce StrandTwister, a novel computational method for detecting beta strand traces in cryo-EM density maps.
• To address the limitation of current tools in identifying beta strands at medium resolutions.
• To enhance the accuracy of atomic structure modeling from cryo-EM data.

Main Methods:

• StrandTwister analyzes the intrinsic twist of beta sheets to identify beta strand pathways.
• The method was tested on simulated beta sheets at 10 Å resolution.
• Validation was performed using beta sheets computationally detected from experimental cryo-EM maps (4.4-7.4 Å resolution).

Main Results:

• StrandTwister successfully detected beta strand traces in simulated and experimental cryo-EM data.
• In experimental maps, StrandTwister detected 81.87% of beta strands with an average distance of 1.66 Å.
• The method demonstrated high accuracy in identifying beta strands, especially in the central regions of beta sheets.

Conclusions:

• StrandTwister is an effective tool for detecting beta strand traces from medium-resolution cryo-EM maps.
• The method significantly improves the ability to model atomic structures by identifying a previously challenging feature.
• StrandTwister offers accurate and reliable detection of beta strands, advancing cryo-EM structure determination.