This document is an erratum for a previously published research paper. While it does not present new experimental data, it provides a critical mathematical correction to a foundational formula used in protein crystallography.
TL;DR
This erratum corrects a typographical error in a key probability equation (Equation 18) from a 2004 paper regarding SAD phasing. It also adds missing funding acknowledgments for the original research.
What Problem Does This Paper Address?
In X-ray crystallography, the "phase problem" is the central challenge: detectors measure the intensity of diffracted beams but lose the phase information required to reconstruct the 3D structure of a molecule.
The original paper by Wang et al. (2004) proposed a hybrid approach to solve this. It combined Direct Methods (mathematical techniques that exploit probabilistic relationships between reflections) with the SOLVE/RESOLVE software suite (automated procedures for finding heavy-atom sites and performing density modification).
The specific problem addressed by this erratum is a mistake in the mathematical expression of the probability distribution used to refine phase estimates. Without the correct formula, researchers attempting to implement or replicate the algorithm would encounter inconsistencies in how anomalous scattering data is integrated with structural triplets.
Methods
The erratum provides the corrected version of Equation (18). This equation calculates the probability $P_+(\Delta \varphi_{\mathbf{h}})$, which is essential for determining the phase doublet in Single-wavelength Anomalous Diffraction (SAD).
The corrected formula utilizes a hyperbolic tangent ($\tanh$) function to weight different sources of phase information:
- Structural Triplets: The summation term $\sum_{\mathbf{h}'} m_{\mathbf{h}'} m_{\mathbf{h}-\mathbf{h}'}$ represents the contribution from direct methods, relating the phases of different reflections.
- Anomalous Signal: The term $x \sin \delta_{\mathbf{h}}$ accounts for the contribution from anomalous scatterers (like selenium or heavy metals) within the crystal.
- Phase Estimates: It incorporates "best" phase values ($\Delta \varphi_{\text{best}}$) derived from previous iterations of the SOLVE/RESOLVE process.
By combining these, the algorithm produces a more accurate probability distribution for the unknown phases, leading to better electron density maps.
Datasets
Not discussed in the paper. (As an erratum, it refers back to the datasets used in the original 2004 publication, which typically included standard protein diffraction benchmarks like vancomycin or insulin).
Key Insights and Ideas
- Mathematical Precision Matters: In probabilistic phasing, even a small typo in a $\tanh$ argument can lead to significant errors in electron density reconstruction.
- Hybridization is Key: The core idea of the original work remains valid: combining the "ab initio" strengths of Direct Methods with the robust, automated environment of SOLVE/RESOLVE creates a more powerful tool for structural biology than either method alone.
- Integration of Anomalous Data: The corrected equation shows exactly how the anomalous scattering component ($x \sin \delta_{\mathbf{h}}$) is additive within the argument of the hyperbolic tangent, acting as a "prior" that shifts the probability based on experimental diffraction differences.
Key Results
The primary result is the corrected Equation (18):
$$P_+(\Delta \varphi_{\mathbf{h}}) = \frac{1}{2} + \frac{1}{2} \tanh \left\{ \sin |\Delta \varphi_{\mathbf{h}}| \left[ \sum_{\mathbf{h}'} m_{\mathbf{h}'} m_{\mathbf{h}-\mathbf{h}'} \times \kappa \sin (\Phi'_3 + \Delta \varphi_{\mathbf{h}'\,\text{best}} + \Delta \varphi_{\mathbf{h}-\mathbf{h}'\,\text{best}}) + x \sin \delta_{\mathbf{h}} \right] \right\}$$
Additionally, the erratum formally acknowledges the support of the Innovation Project of the Chinese Academy of Sciences and the 973 Project from the Ministry of Science and Technology of China.
Limitations and Future Work
As this is an erratum, it does not propose new future work. However, the correction ensures that the software implementations based on this theory (often found in the CCP4 or PHENIX suites) remain mathematically sound. The limitation of this specific formula is that it is tailored for SAD phasing; different formulations are required for Multi-wavelength Anomalous Diffraction (MAD) or Isomorphous Replacement (SIR/MIR).
Summary updated