Improved efficiency in cryo-EM secondary structure topology determination from inaccurate data

Area of Science:

• Structural Biology
• Computational Biology
• Biophysics

Background:

• Determining secondary structure topology is crucial for atomic structure resolution from electron cryo-microscopy (EM) data.
• This process involves matching medium-resolution secondary structures from EM maps with predictions from amino acid sequences.
• Inaccuracies in both data sources necessitate sampling multiple possibilities, increasing computational load.

Purpose of the Study:

• To reduce the computational cost of mapping secondary structures when prediction inaccuracies are considered.
• To develop a more accurate method for identifying secondary structure topology in protein EM studies.

Main Methods:

• A novel method combining dynamic graphs with constrained shortest path algorithms was developed.
• This approach explicitly samples potential secondary structure errors, rather than relying on a single consensus prediction.

Main Results:

• A 34.55% reduction in run-time was achieved compared to conventional methods.
• Improved accuracy was observed when explicitly sampling potential secondary structure errors versus using a consensus prediction.

Conclusions:

• The proposed framework offers a computationally effective approach for identifying optimal secondary structure topology.
• It addresses the challenge of data inaccuracy in protein structure determination from EM.
• This work highlights the potential for developing exact algorithms for complex structural biology problems.