Two-samples t-Tests - Revision history

Laschp: /* Output of the univariate t-test series */

2026-05-01T15:30:17Z

Output of the univariate t-test series

← Older revision		Revision as of 17:30, 1 May 2026
Line 34:		Line 34:
	{\|		{\|
	\|- style="vertical-align:top;"		\|- style="vertical-align:top;"
	\|[[File:t-test-cmdln-output.png\|474px\|thumb\|Command line output of t-tests. In ~~the given~~ example, the m/z segment centered around 2518.0283 Th ~~has~~ the ~~highest~~ potential ~~to discriminate~~ between classes I and II.]]		\|[[File:t-test-cmdln-output.png\|474px\|thumb\|Command line output of t-tests. In this example, the m/z segment centered around 2518.0283 Th shows the greatest potential for distinguishing between classes I and II.]]
	\|[[File:T-test-plot.jpg\|400px\|thumb\|Plot of p-values (log scaled) obtained by univariate t-tests using peak data from individual m/z segments as inputs: The smaller the p-value, the higher the discriminative potential of biomarker peaks at the specific m/z positions]]		\|[[File:T-test-plot.jpg\|400px\|thumb\|Plot of p-values (log scaled) obtained by univariate t-tests using peak data from individual m/z segments as inputs: The smaller the p-value, the higher the discriminative potential of biomarker peaks at the specific m/z positions]]
	\|}		\|}

Laschp: /* Output of the univariate t-test series */

2026-05-01T15:28:43Z

Output of the univariate t-test series

← Older revision		Revision as of 17:28, 1 May 2026
Line 34:		Line 34:
	{\|		{\|
	\|- style="vertical-align:top;"		\|- style="vertical-align:top;"
	\|[[File:t-test-cmdln-output.png\|474px\|thumb\|Command line output of t-tests. In the given example, the m/z segment centered around 2518.0283 Th has the highest potential to discriminate classes I and II.]]		\|[[File:t-test-cmdln-output.png\|474px\|thumb\|Command line output of t-tests. In the given example, the m/z segment centered around 2518.0283 Th has the highest potential to discriminate between classes I and II.]]
	\|[[File:T-test-plot.jpg\|400px\|thumb\|Plot of p-values (log scaled) obtained by univariate t-tests using peak data from individual m/z segments as inputs: The smaller the p-value, the higher the discriminative potential of biomarker peaks at the specific m/z positions]]		\|[[File:T-test-plot.jpg\|400px\|thumb\|Plot of p-values (log scaled) obtained by univariate t-tests using peak data from individual m/z segments as inputs: The smaller the p-value, the higher the discriminative potential of biomarker peaks at the specific m/z positions]]
	\|}		\|}

Laschp: /* Output of the univariate t-test series */

2026-05-01T15:26:37Z

Output of the univariate t-test series

← Older revision		Revision as of 17:26, 1 May 2026
Line 34:		Line 34:
	{\|		{\|
	\|- style="vertical-align:top;"		\|- style="vertical-align:top;"
	\|[[File:t-test-cmdln-output.png\|474px\|thumb\|Command line output of t-tests]]		\|[[File:t-test-cmdln-output.png\|474px\|thumb\|Command line output of t-tests. In the given example, the m/z segment centered around 2518.0283 Th has the highest potential to discriminate classes I and II.]]
	\|[[File:T-test-plot.jpg\|400px\|thumb\|Plot of p-values (log scaled) obtained by univariate t-tests using peak data from individual m/z segments as inputs: The smaller the p-value, the higher the discriminative potential of biomarker peaks at the specific m/z positions]]		\|[[File:T-test-plot.jpg\|400px\|thumb\|Plot of p-values (log scaled) obtained by univariate t-tests using peak data from individual m/z segments as inputs: The smaller the p-value, the higher the discriminative potential of biomarker peaks at the specific m/z positions]]
	\|}		\|}

Laschp: /* Parameter of two-samples t-tests */

2026-05-01T15:20:10Z

Parameter of two-samples t-tests

← Older revision		Revision as of 17:20, 1 May 2026
Line 10:		Line 10:
	== Parameter of two-samples t-tests ==		== Parameter of two-samples t-tests ==

	* m/z range: lower and upper bounds of the m/z region in which the series of t-tests are to be performed		* '''m/z range''': lower and upper bounds of the m/z region in which the series of t-tests are to be performed
	* α: significance level of the t-tests		* '''α''': significance level of the t-tests
	* ''dx'' (ppm): a parameter defining the relative width and thus the number of the spectrum segments. A spectrum segment centered at position <math forcemathmode="png">{x_{i}}</math> covers a m/z interval of an absolute width equaling <math forcemathmode="png">width_{abs,x(i)} = {x_{i}dx10^{-6}}</math>. The lower and upper bounds of the spectrum segments are defined by <math forcemathmode="png">[{x_{i}(1-dx)0.510^{-6}}]</math> (lower bound) and <math forcemathmode="png">[{x_{i}(1+dx)0.510^{-6}}]</math> (upper bound), respectively. Consequently, a spectrum segment of width ''dx''=1000 and centered at <math forcemathmode="png">{x_{i}}</math>=2000 Th would be 2 Th wide with boundaries located at 1999 and 2001 Th		* '''''dx''''' (ppm): a parameter defining the relative width and thus the number of the spectrum segments. A spectrum segment centered at position <math forcemathmode="png">{x_{i}}</math> covers a m/z interval of an absolute width equaling <math forcemathmode="png">width_{abs,x(i)} = {x_{i}dx10^{-6}}</math>. The lower and upper bounds of the spectrum segments are defined by <math forcemathmode="png">[{x_{i}(1-dx)0.510^{-6}}]</math> (lower bound) and <math forcemathmode="png">[{x_{i}(1+dx)0.510^{-6}}]</math> (upper bound), respectively. Consequently, a spectrum segment of width ''dx''=1000 and centered at <math forcemathmode="png">{x_{i}}</math>=2000 Th would be 2 Th wide with boundaries located at 1999 and 2001 Th
	* ~~intensity~~: defines if barcode spectra (checkbox unchecked) or peak weighting factors (checked) are utilized as test inputs		* '''use intensities''': defines if barcode spectra (checkbox unchecked) or peak weighting factors (checked) are utilized as test inputs
	* show histogram: shows a histogram with test outputs (p-values, AUC, etc.) and provides also the mean, median and the standard deviation of the test variables		* '''show histogram''': shows a histogram with test outputs (p-values, AUC, etc.) and provides also the mean, median and the standard deviation of the test variables

	== Performing t-test series ==		== Performing t-test series ==

Laschp: /* Output of the univariate t-test series */

2026-05-01T15:13:39Z

Output of the univariate t-test series

← Older revision		Revision as of 17:13, 1 May 2026
Line 35:		Line 35:
	\|- style="vertical-align:top;"		\|- style="vertical-align:top;"
	\|[[File:t-test-cmdln-output.png\|474px\|thumb\|Command line output of t-tests]]		\|[[File:t-test-cmdln-output.png\|474px\|thumb\|Command line output of t-tests]]
	\|[[File:T-test-plot.jpg\|400px\|thumb\|p-values ~~plot of~~ univariate t-tests ~~(log scaled)~~: The smaller the p-value ~~at the specific m/z position~~ the higher the discriminative potential of biomarker peaks at ~~this position~~]]		\|[[File:T-test-plot.jpg\|400px\|thumb\|Plot of p-values (log scaled) obtained by univariate t-tests using peak data from individual m/z segments as inputs: The smaller the p-value, the higher the discriminative potential of biomarker peaks at the specific m/z positions]]
	\|}		\|}

Laschp: /* Introduction */

2026-05-01T14:58:19Z

Introduction

← Older revision		Revision as of 16:58, 1 May 2026
Line 4:		Line 4:
	[[File:T-test.jpg\|right\|thumb\|Input parameters for univariate t-tests]]		[[File:T-test.jpg\|right\|thumb\|Input parameters for univariate t-tests]]

	This function performs consecutively two-~~samples~~ t-tests in segments of the MALDI-ToF mass spectra using peak table data as inputs. t-tests are useful for biomarker screening in ensembles of labeled MALDI-ToF mass spectra that exhibit certain degree of similarity. A two-~~samples~~ t-test in a given m/z segment returns a test decision for the null hypothesis <math forcemathmode="png">H_{(0)}</math> that the peak intensity data in classes I and II arise from independent random samples from normal distributions with equal means and variances. The alternative hypothesis <math forcemathmode="png">H_{(1)}</math> is that the peak intensity data come from populations with unequal means.		This function performs consecutively two-sample t-tests in segments of the MALDI-ToF mass spectra using peak table data as inputs. t-tests are useful for biomarker screening in ensembles of labeled MALDI-ToF mass spectra that exhibit certain degree of similarity. A two-sample t-test in a given m/z segment returns a test decision for the null hypothesis <math forcemathmode="png">H_{(0)}</math> that the peak intensity data in classes I and II arise from independent random samples from normal distributions with equal means and variances. The alternative hypothesis <math forcemathmode="png">H_{(1)}</math> is that the peak intensity data come from populations with unequal means.

	See also [https://en.wikipedia.org/wiki/Student's_t-test t-test] (Wikipedia)		See also [https://en.wikipedia.org/wiki/Student's_t-test t-test] (Wikipedia)

Laschp: /* Introduction */

2026-05-01T14:56:51Z

Introduction

← Older revision		Revision as of 16:56, 1 May 2026
Line 5:		Line 5:

	This function performs consecutively two-samples t-tests in segments of the MALDI-ToF mass spectra using peak table data as inputs. t-tests are useful for biomarker screening in ensembles of labeled MALDI-ToF mass spectra that exhibit certain degree of similarity. A two-samples t-test in a given m/z segment returns a test decision for the null hypothesis <math forcemathmode="png">H_{(0)}</math> that the peak intensity data in classes I and II arise from independent random samples from normal distributions with equal means and variances. The alternative hypothesis <math forcemathmode="png">H_{(1)}</math> is that the peak intensity data come from populations with unequal means.		This function performs consecutively two-samples t-tests in segments of the MALDI-ToF mass spectra using peak table data as inputs. t-tests are useful for biomarker screening in ensembles of labeled MALDI-ToF mass spectra that exhibit certain degree of similarity. A two-samples t-test in a given m/z segment returns a test decision for the null hypothesis <math forcemathmode="png">H_{(0)}</math> that the peak intensity data in classes I and II arise from independent random samples from normal distributions with equal means and variances. The alternative hypothesis <math forcemathmode="png">H_{(1)}</math> is that the peak intensity data come from populations with unequal means.


	See also [https://en.wikipedia.org/wiki/Student's_t-test t-test] (Wikipedia)		See also [https://en.wikipedia.org/wiki/Student's_t-test t-test] (Wikipedia)

Laschp: /* Introduction */

2026-05-01T14:56:22Z

Introduction

← Older revision		Revision as of 16:56, 1 May 2026
Line 4:		Line 4:
	[[File:T-test.jpg\|right\|thumb\|Input parameters for univariate t-tests]]		[[File:T-test.jpg\|right\|thumb\|Input parameters for univariate t-tests]]

	This function performs consecutively two-samples t-tests in segments of the MALDI-ToF mass spectra using peak table data as inputs. t-tests are useful for biomarker screening in ensembles of labeled MALDI-ToF mass spectra that exhibit certain degree of similarity. A two-samples t-test in a given m/z segment returns a test decision for the null hypothesis <math forcemathmode="png">H_{(0)}</math> that the peak intensity data in classes I and ~~class~~ II arise from independent random samples from normal distributions with equal means and variances. The alternative hypothesis H(1) is that the peak intensity data come from populations with unequal means.		This function performs consecutively two-samples t-tests in segments of the MALDI-ToF mass spectra using peak table data as inputs. t-tests are useful for biomarker screening in ensembles of labeled MALDI-ToF mass spectra that exhibit certain degree of similarity. A two-samples t-test in a given m/z segment returns a test decision for the null hypothesis <math forcemathmode="png">H_{(0)}</math> that the peak intensity data in classes I and II arise from independent random samples from normal distributions with equal means and variances. The alternative hypothesis <math forcemathmode="png">H_{(1)}</math> is that the peak intensity data come from populations with unequal means.

Laschp: /* Introduction */

2026-05-01T14:55:39Z

Introduction

← Older revision		Revision as of 16:55, 1 May 2026
Line 4:		Line 4:
	[[File:T-test.jpg\|right\|thumb\|Input parameters for univariate t-tests]]		[[File:T-test.jpg\|right\|thumb\|Input parameters for univariate t-tests]]

	This function performs two-samples t-tests in ~~each segment~~ of the MALDI-ToF mass spectra using peak table data as inputs. t-tests are useful for biomarker screening in ensembles of mass spectra ~~exhibiting~~ certain degree of similarity. A two-samples t-test in a given m/z segment returns a test decision for the null hypothesis H(0) that the peak intensity data in ~~class~~ I and class II ~~come~~ from independent random samples from normal distributions with equal means and variances. The alternative hypothesis H(1) is that the peak intensity data come from populations with unequal means.		This function performs consecutively two-samples t-tests in segments of the MALDI-ToF mass spectra using peak table data as inputs. t-tests are useful for biomarker screening in ensembles of labeled MALDI-ToF mass spectra that exhibit certain degree of similarity. A two-samples t-test in a given m/z segment returns a test decision for the null hypothesis <math forcemathmode="png">H_{(0)}</math> that the peak intensity data in classes I and class II arise from independent random samples from normal distributions with equal means and variances. The alternative hypothesis H(1) is that the peak intensity data come from populations with unequal means.


	See also [https://en.wikipedia.org/wiki/Student's_t-test t-test] (Wikipedia)		See also [https://en.wikipedia.org/wiki/Student's_t-test t-test] (Wikipedia)

Laschp: /* Parameter of two-samples t-tests */

2026-05-01T14:50:26Z

Parameter of two-samples t-tests

← Older revision		Revision as of 16:50, 1 May 2026
Line 10:		Line 10:
	== Parameter of two-samples t-tests ==		== Parameter of two-samples t-tests ==

	* m/z range: lower and upper bounds of the m/z region in which the t-tests are to be performed		* m/z range: lower and upper bounds of the m/z region in which the series of t-tests are to be performed
	* α: significance level of the t-tests		* α: significance level of the t-tests
	* ''dx'' (ppm): a parameter defining the relative width and thus the number of the spectrum segments. A spectrum segment centered at position <math forcemathmode="png">{x_{i}}</math> covers a m/z interval of an absolute width equaling <math forcemathmode="png">width_{abs,x(i)} = {x_{i}dx10^{-6}}</math>. The lower and upper bounds of the spectrum segments are defined by <math forcemathmode="png">[{x_{i}(1-dx)0.510^{-6}}]</math> (lower bound) and <math forcemathmode="png">[{x_{i}(1+dx)0.510^{-6}}]</math> (upper bound), respectively. Consequently, a spectrum segment of width ''dx''=1000 and centered at <math forcemathmode="png">{x_{i}}</math>=2000 Th would be 2 Th wide with boundaries located at 1999 and 2001 Th		* ''dx'' (ppm): a parameter defining the relative width and thus the number of the spectrum segments. A spectrum segment centered at position <math forcemathmode="png">{x_{i}}</math> covers a m/z interval of an absolute width equaling <math forcemathmode="png">width_{abs,x(i)} = {x_{i}dx10^{-6}}</math>. The lower and upper bounds of the spectrum segments are defined by <math forcemathmode="png">[{x_{i}(1-dx)0.510^{-6}}]</math> (lower bound) and <math forcemathmode="png">[{x_{i}(1+dx)0.510^{-6}}]</math> (upper bound), respectively. Consequently, a spectrum segment of width ''dx''=1000 and centered at <math forcemathmode="png">{x_{i}}</math>=2000 Th would be 2 Th wide with boundaries located at 1999 and 2001 Th