Likelihood-based estimation of substructure content from single-wavelength anomalous diffraction (SAD) intensity data

Area of Science:

• Structural Biology
• Crystallography
• Biophysics

Background:

• Single-wavelength anomalous diffraction (SAD) phasing is crucial for determining protein structures.
• Low signal-to-noise ratios in experimental data present a significant challenge for SAD phasing.
• Accurate estimation of substructure content is vital for successful SAD phasing outcomes.

Purpose of the Study:

• To develop a novel method for accurately estimating the anomalous scattering contribution from measured intensities.
• To provide a reliable measure of the intrinsic anomalous signal in SAD experiments.
• To improve the accuracy of substructure determination in challenging low signal-to-noise conditions.

Main Methods:

• Development of a likelihood-based target function to estimate anomalous scattering strength.
• Determination of a complex correlation parameter relating Bijvoet mates as a function of resolution.
• Accounting for correlated errors in intensity measurements, including those from radiation damage.

Main Results:

• The proposed method provides a novel measure of the intrinsic anomalous signal.
• Refined complex correlation parameters can be interpreted in terms of atomic content before substructure knowledge.
• Maximum-likelihood estimation of substructure content showed high correlation (R=0.72) with phasing results in 357 SAD cases.

Conclusions:

• The developed likelihood-based target function accurately estimates anomalous signal strength.
• This method enhances the reliability of substructure determination in SAD phasing, especially in low SNR scenarios.
• The approach offers a robust way to assess data quality and predict phasing success.