Package 'NPARC'

Title: Non-parametric analysis of response curves for thermal proteome profiling experiments
Description: Perform non-parametric analysis of response curves as described by Childs, Bach, Franken et al. (2019): Non-parametric analysis of thermal proteome profiles reveals novel drug-binding proteins.
Authors: Dorothee Childs, Nils Kurzawa
Maintainer: Nils Kurzawa <[email protected]>
License: GPL-3
Version: 1.17.0
Built: 2024-09-28 03:44:48 UTC
Source: https://github.com/bioc/NPARC

Help Index

Fit sigmoid model

Description

Fit sigmoid model

Usage

fitSingleSigmoid(x, y, start = c(Pl = 0, a = 550, b = 10))

Arguments

x

numeric vector of the independent variables (typically temperature)
y

numeric vector of the dependent variables (typically relative abundance measurements)
start

numeric vector of start parameters for the melting curve equation

Details

Fits the following function to the data: y=(1−Pl)/(1+exp((b−a/x)))+Ply = (1 - Pl) / (1+exp((b - a/x))) + Pl

Value

model summary of type "nls"

Examples

data(stauro_TPP_data_tidy)
stk4 <- dplyr::filter(stauro_TPP_data_tidy, grepl("STK4", uniqueID))
fitSingleSigmoid(stk4$temperature, stk4$relAbundance)

Control parameters for model fitting

Description

Control parameters for model fitting

Usage

getParams(start = c(Pl = 0, a = 550, b = 10), maxAttempts = 100)

Arguments

start

Numeric vector of start parameters for the melting curve equation
maxAttempts

Number of resampling steps in case of unsuccessful model fits

Value

list of two elements: 1) "start" listing the starting parameters for melting curve fitting, 2) "maxAttempts" listing the maximal number of attempts the fit should be allowed

Examples

data(stauro_TPP_data_tidy)
df <- dplyr::filter(stauro_TPP_data_tidy, grepl("MAPK|ATP|CDK|GTP|CRK", uniqueID))
testResults <- runNPARC(x = df$temperature,
                     y = df$relAbundance,
                     id = df$uniqueID,
                     groupsAlt = df$compoundConcentration,
                     dfType = "empirical",
                     control = getParams(maxAttempts = 50))

NPARC package

Description

Non-parametric analysis of response curves

Details

See the preprint on Childs, Bach, Franken et al. (2019): Non-parametric analysis of thermal proteome profiles reveals novel drug-binding proteins

Fit null and alternative models for Non-parametric analysis of response curves

Description

Fit melting curve and return model metrics as well as predictions for the null and alternative models.

Usage

NPARCfit(x, y, id, control = getParams(), groupsNull = NULL, groupsAlt,
  BPPARAM = BiocParallel::SerialParam(progressbar = TRUE),
  returnModels = FALSE)

Arguments

x

numeric vector of the independent variables (typically temperature)
y

numeric vector of the dependent variables (typically relative abundance measurements)
id

character vector with the protein ID to which each each data point belongs.
control

list of parameters used to control specific parts of the analyse
groupsNull

one or more vectors with grouping variables for the null models. See details.
groupsAlt

one or more vectors with grouping variables for the alternative models. See details.
BPPARAM

BiocParallel parameter object to invoke curve fitting in parallel. Default: BiocParallel::SerialParam()
returnModels

boolean value. If true, the fitted models are returned together with the test results

Details

groupsNull or groupsAlt can either be a single vector each, or data.frames of the same length as x and y with one column per factor

Value

data frame with fitted model parameters and additional columns listing e.g. residuals sum of squares

Examples

data(stauro_TPP_data_tidy)
df <- dplyr::filter(stauro_TPP_data_tidy, grepl("CDK|GTP|CRK", uniqueID))
testResults <- NPARCfit(x = df$temperature,
                     y = df$relAbundance,
                     id = df$uniqueID,
                     groupsAlt = df$compoundConcentration)

Perform F-test

Description

Perform F-test

Usage

NPARCtest(modelMetrics, dfType = c("empirical", "theoretical"))

Arguments

modelMetrics

data.frame with results of the model fit in long format.
dfType

character value indicating the method for degrees of freedom computation for the F-test. Theoretical yields the text-book solution. Empirical yields estimates derived from the distribution moments of the RSS.

Value

data frame with fitted model parameters and additional columns listing e.g. residuals sum of squares of null and alterantive model and raw and adjusted p values retrieved from testing

Examples

data(stauro_TPP_data_tidy)
df <- dplyr::filter(stauro_TPP_data_tidy, grepl("CDK|GTP|CRK", uniqueID))
fits <- NPARCfit(x = df$temperature, 
                 y = df$relAbundance, 
                 id = df$uniqueID, 
                 groupsNull = NULL, 
                 groupsAlt = df$compoundConcentration, 
                 returnModels = FALSE)
modelMetrics <- fits$metrics
testRes <-  NPARCtest(modelMetrics, dfType = "theoretical")

Non-parametric analysis of response curves

Description

Wrapper function for melting curve fitting and hypothesis testing.

Usage

runNPARC(x, y, id, groupsNull = NULL, groupsAlt,
  BPPARAM = BiocParallel::SerialParam(progressbar = TRUE),
  dfType = c("theoretical", "empirical"), control = getParams())

Arguments

x

numeric vector of the independent variables (typically temperature)
y

numeric vector of the dependent variables (typically relative abundance measurements)
id

character vector with the protein ID to which each each data point belongs.
groupsNull

one or more vectors with grouping variables for the null models. See details.
groupsAlt

one or more vectors with grouping variables for the alternative models. See details.
BPPARAM

BiocParallel parameter object to invoke curve fitting in parallel. Default: BiocParallel::SerialParam()
dfType

character value indicating the method for degrees of freedom computation for the F-test. Theoretical yields the text-book solution. Empirical yields estimates derived from the distribution moments of the RSS.
control

list of parameters used to control specific parts of the analyse

Details

groupsNull or groupsAlt can either be a single vector each, or data.frames of the same length as x and y with one column per factor

Value

data frame with fitted model parameters and additional columns listing e.g. residuals sum of squares of null and alterantive model

Examples

data(stauro_TPP_data_tidy)
df <- dplyr::filter(stauro_TPP_data_tidy, grepl("CDK|GTP|CRK", uniqueID))
testResults <- runNPARC(x = df$temperature,
                     y = df$relAbundance,
                     id = df$uniqueID,
                     groupsAlt = df$compoundConcentration,
                     dfType = "empirical")

TPP dataset of staurosporine treated cells.

Description

Data from a thermal proteome profiling (TPP) experiment investigating the ATP-competitive pan-kinase inhibitor staurosporine on K562 cells. The data has been downloaded the data from the supplement of the respective publication and converted into tidy format.

Usage

data(stauro_TPP_data_tidy)

Format

An object of class "data.frame"

References

Savitski et al. (2014): Tracking cancer drugs in living cells by thermal profiling of the proteome. Science 346, 1255784.