ConfidenceScoring::GLM_ Struct Reference

Binomial GLM used to map (squared-normalised-RT-diff, intensity-distance) -> [0, 1] confidence. More...

#include <OpenMS/ANALYSIS/OPENSWATH/ConfidenceScoring.h>

Collaboration diagram for ConfidenceScoring::GLM_:

Public Member Functions
double	operator() (double diff_rt, double dist_int) const
	Evaluate the GLM at (diff_rt, dist_int); returns a probability in [0, 1].

Public Attributes
double	intercept
	GLM intercept term.
double	rt_coef
	GLM coefficient on the squared RT difference (units: 1/RT²)
double	int_coef
	GLM coefficient on the Manhattan intensity distance.

Detailed Description

Binomial GLM used to map (squared-normalised-RT-diff, intensity-distance) -> [0, 1] confidence.

Implements \(\mathrm{sigmoid}(\mathrm{intercept} + \mathrm{rt\_coef} \cdot \Delta\mathrm{RT}^2 + \mathrm{int\_coef} \cdot d_\mathrm{int})\) with coefficients fitted externally (see initializeGlm).

Member Function Documentation

operator()()

double operator() ( double  diff_rt, double  dist_int  ) const

inline

Evaluate the GLM at (diff_rt, dist_int); returns a probability in [0, 1].

Member Data Documentation

int_coef

double int_coef

GLM coefficient on the Manhattan intensity distance.

intercept

double intercept

GLM intercept term.

rt_coef

double rt_coef

GLM coefficient on the squared RT difference (units: 1/RT²)