The article discusses the development of an automated method for solving macromolecular structures using multiple isomorphous replacement (MIR) and multiwavelength anomalous scattering (MAD) data. The key challenge in structure determination is the subjective evaluation of heavy-atom partial structures, which is time-consuming and difficult. The authors propose a set of criteria to evaluate the quality of these partial structures, enabling the structure-solution process to be converted into an optimization problem and thus automated. The SOLVE software has been successfully used to solve MAD data sets with up to 52 selenium sites in the asymmetric unit. The process of solving structures using MIR or MAD data involves identifying possible heavy-atom partial structures and evaluating their quality. This is typically done through manual or semi-automated inspection of difference Patterson functions or direct methods. The authors describe a scoring system that allows the structure-determination process to be fully automated. This system includes criteria such as agreement with the Patterson function, cross-validation difference Fourier analysis, figure of merit of phasing, and distinction between solvent and macromolecule in the native Fourier. The scoring system is based on several criteria, including the match between the Patterson function and the trial solution, the peak heights in cross-validation difference Fourier maps, the figure of merit of phasing, and the distinction between solvent and macromolecule in the native Fourier. These criteria are used to evaluate the quality of heavy-atom partial structures and to determine the best solution. The authors also describe the automated structure-determination process, which includes scaling of X-ray data sets, calculation of Patterson and difference Patterson functions, finding and optimizing the heavy-atom partial structure, and calculating native phases and an electron-density map. The process is divided into several steps, including scaling of X-ray data sets, calculation of Patterson and difference Patterson functions, solving the heavy-atom structure, and calculating native phases. The results show that the scoring system is effective in evaluating the quality of heavy-atom partial structures and in identifying the best solution. The overall Z score is a reasonable measure of the quality of a solution, as it performs well across a range of map qualities. The authors conclude that the development of a comprehensive scoring procedure for heavy-atom partial structures could make the structure-determination process well defined and amenable to automation.The article discusses the development of an automated method for solving macromolecular structures using multiple isomorphous replacement (MIR) and multiwavelength anomalous scattering (MAD) data. The key challenge in structure determination is the subjective evaluation of heavy-atom partial structures, which is time-consuming and difficult. The authors propose a set of criteria to evaluate the quality of these partial structures, enabling the structure-solution process to be converted into an optimization problem and thus automated. The SOLVE software has been successfully used to solve MAD data sets with up to 52 selenium sites in the asymmetric unit. The process of solving structures using MIR or MAD data involves identifying possible heavy-atom partial structures and evaluating their quality. This is typically done through manual or semi-automated inspection of difference Patterson functions or direct methods. The authors describe a scoring system that allows the structure-determination process to be fully automated. This system includes criteria such as agreement with the Patterson function, cross-validation difference Fourier analysis, figure of merit of phasing, and distinction between solvent and macromolecule in the native Fourier. The scoring system is based on several criteria, including the match between the Patterson function and the trial solution, the peak heights in cross-validation difference Fourier maps, the figure of merit of phasing, and the distinction between solvent and macromolecule in the native Fourier. These criteria are used to evaluate the quality of heavy-atom partial structures and to determine the best solution. The authors also describe the automated structure-determination process, which includes scaling of X-ray data sets, calculation of Patterson and difference Patterson functions, finding and optimizing the heavy-atom partial structure, and calculating native phases and an electron-density map. The process is divided into several steps, including scaling of X-ray data sets, calculation of Patterson and difference Patterson functions, solving the heavy-atom structure, and calculating native phases. The results show that the scoring system is effective in evaluating the quality of heavy-atom partial structures and in identifying the best solution. The overall Z score is a reasonable measure of the quality of a solution, as it performs well across a range of map qualities. The authors conclude that the development of a comprehensive scoring procedure for heavy-atom partial structures could make the structure-determination process well defined and amenable to automation.