ALIXE

ALIXE combines information from multiple partial solutions to improve the starting map to reveal the full structure. This has been shown to aid structure solution, especially in difficult cases. Also, combining partial solutions reduces redundancy and hence the time required to evaluate them.

The target structure is characterized by its set of phases. Any correct partial structure should be consistent with this unknown phase set. ALIXE builds the average for each set of consistent solutions. Clustering and origin shift are reevaluated against each combined phase set until convergence. If all fragments in the structure were included, the process should reconstruct the true phase set. In practice, the sets of correctly placed fragments are incomplete and imperfect, but their correct combination should provide a better approximation than any single solution. An inherent complication is that as the fragment placements are not independent in their generation, sets of wrong consistent probes will also be present. The ability to expand from the combined solutions finally establishes the correctness of the hypothesis.

It is integrated into the ARCIMBOLDO programs as well as available as a standalone to enable its use on results from other phasing methods. Other features include reduction of redundancy of solutions in highly repetitive structures, such as coiled coils, and computation of map correlation coefficients to study the statistical relations between solutions with CC_ANALYSIS (Diederichs, 2017).

ALIXE has been optimized allowing its use even with modest hardware, including laptops.

ALIXE ships in the ARCIMBOLDO pip package. See the Download page.