core::scoring::epr_deer::Simulated4PDEERTraceFactory Class Reference

#include <Simulated4PDEERTraceFactory.hh>

Public Member Functions
	Simulated4PDEERTraceFactory ()=default
	Default constructor. More...
	Simulated4PDEERTraceFactory (utility::vector1< Real > const &exp_trace, utility::vector1< Real > const &time_pts, std::string const &bckg_type, Size const &bins_per_a, Size const &max_dist=100, Size const &n_bins=200)
	More comprehensive constructor that gets everything in order. More...
numeric::MathMatrix< Real >	initialize_kernel (utility::vector1< Real > const &time_pts, Size const &bins_per_a, Size const &max_dist=100, Size const &n_bins=200) const
	Kernel matrix initialization method. More...
Simulated4PDEERTrace	trace_from_distr (std::map< Size, Real > const &distr)
	Process for converting distance distribution to DEER trace. More...
utility::vector1< Real >	kernel_mult (std::map< Size, Real > const &distr) const
	Actual nitty-gritty code for applying the kernel matrix. More...
utility::vector1< Real >	opt_bckg (utility::vector1< Real > const &deer_trace)
	Optimize the intermolecular background by least squares. More...
void	initial_search (utility::vector1< Real > const &deer_trace)
	Performs an initial search for background parameters. More...
void	append_data_and_recalculate (Real const &trace_datum, Real const &time_point)
	Adds a time point to the saved experimental DEER data. More...
void	reset ()
	Reset background calculation to force initial search. More...
utility::vector1< Real >	trace () const
	Returns experimental data. More...
utility::vector1< Real >	time_pts () const
	Returns time points for experimental data. More...
std::string	bckg_type () const
	Returns background type. More...
Size	bins_per_a () const
	Returns bins per angstrom in the kernel matrix. More...
Real	depth () const
Real	slope () const
Real	dim () const
void	trace (utility::vector1< Real > const &val)
	Sets experimental data to new DEER trace. More...
void	time_pts (utility::vector1< Real > const &val)
	Sets time points to new data. More...
void	bckg_type (std::string const &val)
	Sets background coupling type. More...
void	bins_per_a (Size const &val)
	Sets bins per angstrom of kernel matrix and recalculates matrix. More...

Private Attributes
utility::vector1< Real >	exp_data_
	Experimental data. More...
utility::vector1< Real >	time_pts_
	Time points. More...
std::string	bckg_type_
	Background type ("3D", "NON-3D", "NONE") More...
Size	bins_per_a_
	Bins per angstrom in kernel matrix. More...
Size	max_dist_
Size	n_bins_ = 200
	Number of bins for magnetic field vector angle calculation. More...
numeric::MathMatrix< Real >	kernel_
	Matrix that converts distribution to intramolecular DEER trace. More...
Real	depth_ = 0.25
	Last saved modulation depth. More...
Real	slope_ = 1e-4
	Last saved background slope. More...
Real	dim_ = dim3d_
	Last saved dimensionality Default value chosen because sigmoid( 0.6931, 2, 3.5 ) = 3 See sigmoid function in util.cc for details. More...
bool	optimized_ = false
	To know if more LM iterations need to be run. More...
Size	init_runs_ = 100
	How many times to run LM the first time. More...
Size	runs_ = 5
	How many times to run LM subsequent times. More...
Real	tpts_sqd_ = 0.0
	Time points squared, used by initial_search() fxn. More...

Constructor & Destructor Documentation

core::scoring::epr_deer::Simulated4PDEERTraceFactory::Simulated4PDEERTraceFactory ( )

default

Default constructor.

core::scoring::epr_deer::Simulated4PDEERTraceFactory::Simulated4PDEERTraceFactory	(	utility::vector1< Real > const &	exp_trace,
		utility::vector1< Real > const &	time_pts,
		std::string const &	bckg_type,
		Size const &	bins_per_a,
		Size const &	max_dist = 100,
		Size const &	n_bins = 200
	)

More comprehensive constructor that gets everything in order.

Parameters

exp_trace	Experimental DEER trace for comparison
time_pts	Time points for the experimental DEER trace
bckg_type	Type of background. Can be "NONE", "3D", "NON-3D"
max_dist	Maximum distance in angstroms
n_bins	For discretization of angle-dependent dipolar coupling

References bckg_type(), bckg_type_, bins_per_a(), bins_per_a_, exp_data_, initialize_kernel(), kernel_, max_dist_, n_bins_, time_pts(), time_pts_, and tpts_sqd_.

Member Function Documentation

void core::scoring::epr_deer::Simulated4PDEERTraceFactory::append_data_and_recalculate	(	Real const &	trace_datum,
		Real const &	time_point
	)

Adds a time point to the saved experimental DEER data.

Parameters

trace_datum	The datapoint (Y-axis)
time_point	The datapoint's time (X-axis) NOT RECOMMENDED. The entire kernel matrix is reinitialized! This is insanely time-consuming!

References bins_per_a_, exp_data_, initialize_kernel(), kernel_, max_dist_, n_bins_, and time_pts_.

std::string core::scoring::epr_deer::Simulated4PDEERTraceFactory::bckg_type

(

)

const

Returns background type.

Returns

Background type

References bckg_type_.

Referenced by Simulated4PDEERTraceFactory().

void core::scoring::epr_deer::Simulated4PDEERTraceFactory::bckg_type

(

std::string const &

val

)

Sets background coupling type.

Parameters

val	New background coupling

References bckg_type_, and protocols::hybridization::val.

Size core::scoring::epr_deer::Simulated4PDEERTraceFactory::bins_per_a

(

)

const

Returns bins per angstrom in the kernel matrix.

Returns

Bins per angstrom in the kernel matrix

References bins_per_a_.

Referenced by Simulated4PDEERTraceFactory().

void core::scoring::epr_deer::Simulated4PDEERTraceFactory::bins_per_a

(

Size const &

val

)

Sets bins per angstrom of kernel matrix and recalculates matrix.

Parameters

val	New value for bins per angstrom

References bins_per_a_, initialize_kernel(), kernel_, max_dist_, n_bins_, time_pts_, and protocols::hybridization::val.

Real core::scoring::epr_deer::Simulated4PDEERTraceFactory::depth

(

)

const

References depth_, core::scoring::epr_deer::maxdepth_, core::scoring::epr_deer::mindepth_, and core::scoring::epr_deer::sigmoid().

Referenced by initial_search().

Real core::scoring::epr_deer::Simulated4PDEERTraceFactory::dim

(

)

const

References dim_, core::scoring::epr_deer::maxdim_, core::scoring::epr_deer::mindim_, and core::scoring::epr_deer::sigmoid().

void core::scoring::epr_deer::Simulated4PDEERTraceFactory::initial_search

(

utility::vector1< Real > const &

deer_trace

)

Performs an initial search for background parameters.

Parameters

deer_trace

Intramolecular DEER trace

References core::scoring::epr_deer::add_bckg(), depth(), depth_, exp_data_, core::scoring::epr_deer::logit(), protocols::mean_field::max(), core::scoring::epr_deer::maxdepth_, core::scoring::epr_deer::mindepth_, protocols::hybridization::score, slope_, protocols::hybridization::t, time_pts_, tpts_sqd_, and protocols::hybridization::val.

Referenced by opt_bckg().

numeric::MathMatrix< Real > core::scoring::epr_deer::Simulated4PDEERTraceFactory::initialize_kernel	(	utility::vector1< Real > const &	time_pts,
		Size const &	bins_per_a,
		Size const &	max_dist = 100,
		Size const &	n_bins = 200
	)		const

Kernel matrix initialization method.

Parameters

time_pts	Time points to be used in the kernel
bins_per_a	Bins per angstrom
max_dist	Maximum distance in Angstroms to be stored here
n_bins	For discretization of angle-dependent dipolar coupling

Returns

Kernel matrix Mathematical details can be found in Ibanez and Jeschke, 2019 JMR (doi: 10.1016/j.jmr.2019.01.008)

References core::pose::motif::a(), core::pack::dunbrack::c, protocols::hybridization::t, and protocols::hybridization::val.

Referenced by append_data_and_recalculate(), bins_per_a(), Simulated4PDEERTraceFactory(), and time_pts().

utility::vector1< Real > core::scoring::epr_deer::Simulated4PDEERTraceFactory::kernel_mult

(

std::map< Size, Real > const &

distr

)

const

Actual nitty-gritty code for applying the kernel matrix.

Parameters

distr

DEER distribution stored as a map

Returns

Vector with intramolecular component Note that throughout this calculation, the bins per angstrom and time points are assumed to be correct! This is not checked (maybe a debug assert can be used?)

References kernel_, core::scoring::epr_deer::TR(), trace(), and protocols::hybridization::val.

Referenced by trace_from_distr().

utility::vector1< Real > core::scoring::epr_deer::Simulated4PDEERTraceFactory::opt_bckg

(

utility::vector1< Real > const &

deer_trace

)

Optimize the intermolecular background by least squares.

Parameters

deer_trace

The intramolecular deer trace

Returns

vector with final DEER trace The function can be run with and without dimensionality calculation. This is reserved for cases where membrane proteins are studied and are not uniformly distributed in three-dimensional space.

References core::scoring::epr_deer::add_bckg(), bckg_type_, depth_, dim_, exp_data_, init_runs_, initial_search(), core::scoring::epr_deer::maxdepth_, core::scoring::epr_deer::mindepth_, core::scoring::epr_deer::minimize_sumsq(), core::scoring::epr_deer::minimize_sumsq_nodim(), optimized_, runs_, core::scoring::epr_deer::sigmoid(), slope_, time_pts_, and core::scoring::epr_deer::TR().

Referenced by trace_from_distr().

void core::scoring::epr_deer::Simulated4PDEERTraceFactory::reset

(

)

Reset background calculation to force initial search.

References optimized_.

Real core::scoring::epr_deer::Simulated4PDEERTraceFactory::slope

(

)

const

References slope_.

utility::vector1< Real > core::scoring::epr_deer::Simulated4PDEERTraceFactory::time_pts

(

)

const

Returns time points for experimental data.

Returns

Time points for experimental data

References time_pts_.

Referenced by Simulated4PDEERTraceFactory().

void core::scoring::epr_deer::Simulated4PDEERTraceFactory::time_pts

(

utility::vector1< Real > const &

val

)

Sets time points to new data.

Sets time points to new data and recalculates matrix.

Parameters

val	New data to set

References bins_per_a_, initialize_kernel(), kernel_, max_dist_, n_bins_, time_pts_, and protocols::hybridization::val.

utility::vector1< Real > core::scoring::epr_deer::Simulated4PDEERTraceFactory::trace

(

)

const

Returns experimental data.

Returns

DEER data

References exp_data_.

Referenced by kernel_mult(), and core::scoring::epr_deer::metrics::DEERDecayData::sum_of_squares().

void core::scoring::epr_deer::Simulated4PDEERTraceFactory::trace

(

utility::vector1< Real > const &

val

)

Sets experimental data to new DEER trace.

Parameters

val	New data to set

References exp_data_, and protocols::hybridization::val.

Simulated4PDEERTrace core::scoring::epr_deer::Simulated4PDEERTraceFactory::trace_from_distr

(

std::map< Size, Real > const &

distr

)

Process for converting distance distribution to DEER trace.

Parameters

distr

DEER distribution stored as a map

Returns

DEER trace with all the bells and whistles attached

References depth_, dim_, kernel_mult(), opt_bckg(), slope_, and time_pts_.

Referenced by core::scoring::epr_deer::metrics::DEERDecayData::get_score().

Member Data Documentation

std::string core::scoring::epr_deer::Simulated4PDEERTraceFactory::bckg_type_

private

Background type ("3D", "NON-3D", "NONE")

Referenced by bckg_type(), opt_bckg(), and Simulated4PDEERTraceFactory().

Size core::scoring::epr_deer::Simulated4PDEERTraceFactory::bins_per_a_

private

Bins per angstrom in kernel matrix.

Referenced by append_data_and_recalculate(), bins_per_a(), Simulated4PDEERTraceFactory(), and time_pts().

Real core::scoring::epr_deer::Simulated4PDEERTraceFactory::depth_ = 0.25

private

Last saved modulation depth.

Referenced by depth(), initial_search(), opt_bckg(), and trace_from_distr().

Real core::scoring::epr_deer::Simulated4PDEERTraceFactory::dim_ = dim3d_

private

Last saved dimensionality Default value chosen because sigmoid( 0.6931, 2, 3.5 ) = 3 See sigmoid function in util.cc for details.

Referenced by dim(), opt_bckg(), and trace_from_distr().

utility::vector1< Real > core::scoring::epr_deer::Simulated4PDEERTraceFactory::exp_data_

private

Experimental data.

Referenced by append_data_and_recalculate(), initial_search(), opt_bckg(), Simulated4PDEERTraceFactory(), and trace().

Size core::scoring::epr_deer::Simulated4PDEERTraceFactory::init_runs_ = 100

private

How many times to run LM the first time.

Referenced by opt_bckg().

numeric::MathMatrix< Real > core::scoring::epr_deer::Simulated4PDEERTraceFactory::kernel_

private

Matrix that converts distribution to intramolecular DEER trace.

Referenced by append_data_and_recalculate(), bins_per_a(), kernel_mult(), Simulated4PDEERTraceFactory(), and time_pts().

Size core::scoring::epr_deer::Simulated4PDEERTraceFactory::max_dist_

private

Referenced by append_data_and_recalculate(), bins_per_a(), Simulated4PDEERTraceFactory(), and time_pts().

Size core::scoring::epr_deer::Simulated4PDEERTraceFactory::n_bins_ = 200

private

Number of bins for magnetic field vector angle calculation.

Referenced by append_data_and_recalculate(), bins_per_a(), Simulated4PDEERTraceFactory(), and time_pts().

bool core::scoring::epr_deer::Simulated4PDEERTraceFactory::optimized_ = false

private

To know if more LM iterations need to be run.

Referenced by opt_bckg(), and reset().

Size core::scoring::epr_deer::Simulated4PDEERTraceFactory::runs_ = 5

private

How many times to run LM subsequent times.

Referenced by opt_bckg().

Real core::scoring::epr_deer::Simulated4PDEERTraceFactory::slope_ = 1e-4

private

Last saved background slope.

Referenced by initial_search(), opt_bckg(), slope(), and trace_from_distr().

utility::vector1< Real > core::scoring::epr_deer::Simulated4PDEERTraceFactory::time_pts_

private

Time points.

Referenced by append_data_and_recalculate(), bins_per_a(), initial_search(), opt_bckg(), Simulated4PDEERTraceFactory(), time_pts(), and trace_from_distr().

Real core::scoring::epr_deer::Simulated4PDEERTraceFactory::tpts_sqd_ = 0.0

private

Time points squared, used by initial_search() fxn.

Referenced by initial_search(), and Simulated4PDEERTraceFactory().

---

The documentation for this class was generated from the following files:

• src/core/scoring/epr_deer/Simulated4PDEERTraceFactory.hh
• src/core/scoring/epr_deer/Simulated4PDEERTraceFactory.cc