Determination of molecular envelopes from solvent contrast variation data

Abstract

An algorithm is described for determining macromolecular envelopes from crystal diffraction amplitudes measured from a solvent contrast variation series. The method uses solvent contrast variation data that have been preprocessed to represent the structure-factor amplitudes of the envelope. The amplitudes are phased using an iterative projection algorithm that incorporates connectivity and compactness constraints on the envelope. The algorithm is tested by simulation on two protein envelopes and shown to be effective even in the absence of the very low resolution data, which are difficult to access experimentally.

Direct phasing of protein crystals with high solvent content

“…Partial knowledge of some of the phases, for example, can easily be incorporated in the iteration. Prior real-space information such as solvent region (Zou & Phillips, 1998;Lo et al, 2009) or protein fragments can also be employed in the density modification. The large number of iterations helps to eliminate the bias of initial phases.…”

Iterative projection algorithms in protein crystallography. I. Theory

“…It is therefore not prone to stagnation. As noted above, the difference-map algorithm has been used in a number of crystallographic applications (Elser, 2003b;Lo et al, 2009;Lo & Millane, 2010).…”

“…The hybrid input-output algorithm was applied to phasing X-ray data from single particles (Miao et al, 1999;Chapman et al, 2006). A projection algorithm called the 'difference map algorithm' (Elser, 2003a) has been applied ab initio to small-molecule crystallography (Elser, 2003b), single-particle imaging (Thibault et al, 2006), determination of molecular envelopes from solvent contrast variation data (Lo et al, 2009), and virus crystallography (Lo & Millane, 2010). Most recently, Liu et al (2012) applied the hybrid input-output algorithm to a number of proteins with high (> 65%) solvent content.…”

Iterative projection algorithms for ab initio phasing in virus crystallography

“…Millane and Stroud (1997) [18] and van der Plas and Millane (2000) [28] adapted the hybrid input-output (HIO) algorithm [6] to incorporate a NCS constraint and applied it to reconstruction of an icosahedral virus at medium resolution using simulated data. Lo et al (2009) [9] applied the difference map algorithm to determination of molecular envelopes 150 in protein crystals from simulated solvent contrast variation data. Lo and Millane (2010) [11] applied the difference map algorithm to reconstruction of a previously solved icosahedral virus using experimental data, which gave reasonably good electron density maps.…”