• Manual
  • Introduction
  • Running sasrefmx
    • Command-Line
      • Arguments and Options
      • Interactive Configuration
      • Runtime Output
    • Graphical Interface
  • sasrefmx Input Files
    • Data control file
    • Subunits control file
    • Cross-correlation file
    • Distance restraints
  • sasrefmx Output Files
• Examples
  • Quaternary Structure Analysis of a Weak Tetramer
  • Modelling of a Transient Heterodimer
  • Complex with Excess of a Component in Solution
  • Presence of a Sub-Complex in the Mixture

Manual

The following describes SASREFMX, a special version of the rigid body modelling program SASREF for structural analysis of transient complexes and weak oligomers from polydisperse scattering data.

SASREFMX couples rigid body modelling with mixture analysis, estimating volume fractions of associated states and dissociation products. This manual describes interactive configuration, required input files, and produced output files.

Introduction

SASREFMX performs quaternary structure modelling of a complex particle formed by subunits with known atomic structures against scattering data from polydisperse samples. Multiple curves can be fitted simultaneously, e.g. profiles measured at different concentration, temperature, pH, or ionic strength conditions, reflecting different affinities and association states.

The algorithm can account for symmetry (global, subunit-specific, and curve-specific). A simulated annealing protocol constructs an interconnected ensemble of subunits without steric clashes while minimizing discrepancy between experimental and predicted scattering.

SASREFMX models each curve as a linear combination of the intact complex and predefined dissociation products. These dissociation products are defined implicitly via a dummy entry in the data control file and explicitly in the last row of the cross-correlation matrix.

For further details of the rigid body modelling approach, please refer to the SASREF manual.

Running sasrefmx

Command-Line

Usage:

$ sasrefmx [OPTIONS]

SASREFMX is configured in interactive mode; command-line options are used to supplement the run.

Arguments and Options

SASREFMX recognizes the following command-line options. Mandatory arguments to long options are mandatory for short options too.

Interactive Configuration

SASREFMX runs in dialog mode. A substantial part of the setup is provided through configuration files.

There are two modes, EXPERT and USER. In EXPERT mode, more parameters can be changed. In USER mode, fewer questions are asked and defaults are used for most parameters. The default values are identical in both modes.

An interactive answers file (.ans) may be used to record and replay configurations, enabling repeatable runs without re-entering parameters.

Runtime Output

On runtime, output includes temperature-step progress, best \(\chi^2\) values, and associated volume fractions:

j:   4 T: 0.729E+01 Suc:  1000 Eva:    12497 CPU:  0.208E+03 F:99.4301 Pen: 13.803
The best chi^2 values:11.64871 5.96331
Associated volume fractions: 0.42157 0.64983

The fields can be interpreted as follows:

Graphical Interface

There is no graphical interface for SASREFMX.

sasrefmx Input Files

Three compulsory configuration files are required:

• scattering data
• subunits (rigid bodies)
• subunit-to-curve contribution table

Data control file

• The first row shall indicate the total number \(K\) of rows following
• \(K\) rows follow; each row has 8 whitespace-separated values/columns
• a last dummy line describing dissociation products. It keeps placeholder values in most fields; symmetry is used in downstream construct generation. The cross-correlation file has to account for this.

Values/columns for each data row, in order:

Subunits control file

The subunits control file describes the rigid bodies:

• The first row shall indicate the total number \(M\) of rows following
• \(M\) rows follow; each row has 4 whitespace-separated values/columns

Values/columns for each data row, in order:

Cross-correlation file

The cross-correlation file ties the experimental data and subunits together. It contains a matrix with \(M\) columns and \(K\) rows, where \(M\) is the number of subunits and \(K\) is the number of curves.

Entry (i,j) of this matrix specifies the contribution of the i-th subunit to the j-th data set.

For SANS curves the matrix value ([0.0-1.0 or -1.0]) is the subunit perdeuteration (level of D₂O in expression medium); -1.0 means the subunit is not present in that construct. For X-ray curves, use 0.0 if the subunit is present, -1.0 otherwise.

As the data control file adds a dummy entry for dissociation products, the last row of the cross-correlation file has to account for this:

• -1 means the subunit is absent from dissociation products.
• 0 means the subunit is part of a single sub-complex product.
• positive integer values define stoichiometric coefficients for fully dissociated components.

Distance restraints

Distance restraints may be imposed via an optional contacts conditions file:

dist 7.0
1 0 0 2 1 1
dist 5.0
2 0 0 3 1 1
dist 7.0
1 342 342 2 25 25
1 350 350 2 17 17
dist 6.0
1 290 297 2  64 79
dist 7.0
1 1 0 3 1 0

dist 7.0 means that the minimum distance between CA atoms of selected residue ranges (or P atoms in nucleotides) should not exceed 7 \(\text{\AA}\).

In a line without dist, the 1st and 4th values are ordinal subunit numbers. The 2nd-3rd and 5th-6th values specify residue ranges in first and second subunits, respectively; 0 means the last residue/nucleotide.

If multiple alternatives are listed after one dist line, the smallest alternative distance is compared to the threshold.

Please refer to the SASREF manual for details.

Important (new as of ATSAS-4.0): in the presence of symmetry, subunit numbering differs from older versions. First all symmetry mates of subunit 1 are listed, then of subunit 2, and so on.

sasrefmx Output Files

After each simulated annealing step, SASREFMX writes output files using the log-file prefix. Existing files with the same prefix are overwritten.

Examples

Quaternary Structure Analysis of a Weak Tetramer

Let weak_tetramer.dat be a SAXS profile from a polydisperse sample containing an equilibrium between a P222 tetramer and its monomer. The monomer atomic structure is in monomer.pdb. This can be described by:

Data control file, curves.con:

2
weak_tetramer.dat -1.00 P222 1 1.0 0 1.0 y
dummy.dat         -1.00 P1   1 1.0 0 1.0 y

Subunits control file, subunits.con:

1
monomer.pdb Y N P222

Cross-correlation file, table.con:

0.0
0.0

Modelling of a Transient Heterodimer

An equimolar mixture of proteins A and B yields an equilibrium between AB and unbound A/B. A concentration series with different associated fractions can be fitted simultaneously.

Data control file, curves.con:

4
transient_c1.dat -1.00 P1 1 1.0 0 1.0 y
transient_c2.dat -1.00 P1 1 1.0 0 1.0 y
transient_c3.dat -1.00 P1 1 1.0 0 1.0 y
dummy.dat        -1.00 P1 1 1.0 0 1.0 y

Subunits control file, subunits.con:

2
a.pdb Y N P1
b.pdb Y N P1

Cross-correlation file, table.con:

0.0 0.0
0.0 0.0
0.0 0.0
1.0 1.0

Complex with Excess of a Component in Solution

If excess subunit C is required to form stable complex ABC, the sample can be polydisperse. Configuration files:

Data control file, curves.con:

2
mixture.dat -1.00 P1 1 1.0 0 1.0 y
dummy.dat   -1.00 P1 1 1.0 0 1.0 y

Subunits control file, subunits.con:

3
a.pdb Y N P1
b.pdb Y N P1
c.pdb Y N P1

Cross-correlation file, table.con:

0.0 0.0 0.0
-1.0 -1.0 0.0

Presence of a Sub-Complex in the Mixture

If subunit C is limiting (e.g. stoichiometric study), sub-complex AB may be present together with ternary ABC. Configuration files:

Data control file, curves.con:

2
mixture.dat -1.00 P1 1 1.0 0 1.0 y
dummy.dat   -1.00 P1 1 1.0 0 1.0 y

Subunits control file, subunits.con:

3
a.pdb Y N P1
b.pdb Y N P1
c.pdb Y N P1

Cross-correlation file, table.con:

0.0 0.0 0.0
0.0 0.0 -1.0