An automated data processing and analysis pipeline for transmembrane proteins in detergent solutions

D. Molodenskiy; H. Mertens; D. Svergun

The AMPP is implemented as an extension of SASFLOW (33) , an automated data reduction and analysis pipeline currently running at P12. AMPP utilizes several modules and modeling programs available in the ATSAS package (34) and, in addition, calls the program MEMPROT (28) . The primary data analysis is executed directly upon completion of a SEC-SAXS data acquisition and proceeds in the following workflow ( Figure. 1) Radial integration of 2D SAXS images by RADAVER (33) (typically 1000-3000 frames are collected for a standard SEC-SAXS run set with one second exposure per image/frame). 2) Determination of the buffer and sample ranges from the elution profile using CHROMIXS (35) , an important analysis component of the pipeline specifically designed for SEC-SAXS analysis. CHROMIXS automatically identifies both buffer and sample regions in the elution profile, and performs frame averaging and subtraction to produce the SAXS curve of the purified component for subsequent analysis (Figure 2 ). It is worth noting that CHROMIXS is capable of identifying multiple elution peaks, providing the users with a set of distinct SAXS curves from individual fractions, if present. 3) Calculation of the overall parameters (radius of gyration R g , maximum size D max , excluded volume V Porod , molecular weight MW) and the generation of relevant plots (Guinier plot: ln I(s) vs s 2 ; Kratky plot: I(s)s 2 vs s; and the real-space distance distribution function: P(r) vs r) for all subtracted curves. An example of these plots for the mechanosensitive T2 channel data is shown in the generated summary table (Figure. 3). These parameters are employed for the evaluation of the search volume to conduct the ab initio reconstruction using MONSA or for the hybrid MEMPROT fitting procedures. 4) Querying and collecting the available a priori information from the ISPyB database via dedicated Simple Object Access Protocol (SOAP) web services. Depending on the results, one of the three modeling paths is executed for the subsequent modeling ( Figure. 1):