Type:	Package
Title:	Proteomics Data Analysis and Modeling Tools
Version:	0.2.1
Description:	A comprehensive, user-friendly package for label-free proteomics data analysis and machine learning-based modeling. Data generated from 'MaxQuant' can be easily used to conduct differential expression analysis, build predictive models with top protein candidates, and assess model performance. promor includes a suite of tools for quality control, visualization, missing data imputation (Lazar et. al. (2016) <doi:10.1021/acs.jproteome.5b00981>), differential expression analysis (Ritchie et. al. (2015) <doi:10.1093/nar/gkv007>), and machine learning-based modeling (Kuhn (2008) <doi:10.18637/jss.v028.i05>).
License:	LGPL-2.1 | LGPL-3 [expanded from: LGPL (≥ 2.1)]
Encoding:	UTF-8
Language:	en-US
RoxygenNote:	7.2.2
VignetteBuilder:	knitr
Suggests:	covr, knitr, rmarkdown, testthat (≥ 3.0.0)
Depends:	R (≥ 3.5.0)
URL:	https://github.com/caranathunge/promor, https://caranathunge.github.io/promor/
Imports:	reshape2, ggplot2, ggrepel, gridExtra, limma, statmod, pcaMethods, VIM, missForest, caret, kernlab, xgboost, naivebayes, viridis, pROC
LazyData:	true
Config/testthat/edition:	3
BugReports:	https://github.com/caranathunge/promor/issues
NeedsCompilation:	no
Packaged:	2023-07-17 22:04:15 UTC; caran
Author:	Chathurani Ranathunge [aut, cre, cph]
Maintainer:	Chathurani Ranathunge <caranathunge86@gmail.com>
Repository:	CRAN
Date/Publication:	2023-07-17 22:30:02 UTC

Compute average intensity

Description

This function computes average intensities across technical replicates for each sample.

Usage

aver_techreps(raw_df)

Arguments

Details

aver_techreps assumes that column names in the data frame follow the "Group_UniqueSampleID_TechnicalReplicate" notation. (Use head(raw_df) to see the structure of the raw_df object.)

Value

A raw_df object of averaged intensities.

Author(s)

Chathurani Ranathunge

See Also

• avearrays from limma package.

• create_df

Examples

## Use a data set containing technical replicates to create a raw_df object
raw_df <- create_df(
prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg2.txt",
exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed2.txt",
tech_reps = TRUE
)

# Compute average intensities across technical replicates.
rawdf_ave <- aver_techreps(raw_df)

Correlation between technical replicates

Description

This function generates scatter plots to visualize the correlation between a given pair of technical replicates (Eg: 1 vs 2) for each sample.

Usage

corr_plot(
  raw_df,
  rep_1,
  rep_2,
  save = FALSE,
  file_type = "pdf",
  palette = "viridis",
  text_size = 5,
  n_row = 4,
  n_col = 4,
  dpi = 80,
  file_path = NULL
)

Arguments

Details

• Given a data frame of log-transformed intensities (a raw_df object) and a pair of numbers referring to the technical replicates, corr_plot produces a list of scatter plots showing correlation between the given pair of technical replicates for all the samples provided in the data frame.

• Note: n_row * n_col should be equal to the number of samples to display in a single page.

Value

A list of ggplot2 plot objects.

Author(s)

Chathurani Ranathunge

See Also

create_df

Examples

## Use a data set containing technical replicates to create a raw_df object
raw_df <- create_df(
prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg2.txt",
exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed2.txt",
tech_reps = TRUE
)



## Compare technical replicates 1 vs. 2 for all samples
corr_plot(raw_df, rep_1 = 1, rep_2 = 2)

Suvarna et al 2021 LFQ data (fit object)

Description

An object of class "MArrayLM" from running find_dep on covid_norm_df

Usage

data(covid_fit_df)

Format

An object of class "MArrayLM"

References

https://www.frontiersin.org/articles/10.3389/fphys.2021.652799/full#h3

Suvarna et al 2021 LFQ data (normalized)

Description

A dataframe containing normalized LFQ protein intensity data for 230 proteins in 35 samples (a subset of the original data set)

Usage

data(covid_norm_df)

Format

A data frame with 230 rows (proteins) and 35 columns (samples)

References

https://www.frontiersin.org/articles/10.3389/fphys.2021.652799/full#h3

Create a data frame of protein intensities

Description

This function creates a data frame of protein intensities

Usage

create_df(
  prot_groups,
  exp_design,
  input_type = "MaxQuant",
  data_type = "LFQ",
  filter_na = TRUE,
  filter_prot = TRUE,
  uniq_pep = 2,
  tech_reps = FALSE,
  zero_na = TRUE,
  log_tr = TRUE,
  base = 2
)

Arguments

Details

• This function first reads in the proteinGroups.txt file produced by MaxQuant or a standard input file containing a quantitative matrix where the proteins or protein groups are indicated by rows and the samples by columns.

• It then reads in the expDesign.txt file provided as exp_design and extracts relevant information from it to add to the data frame. an example of the expDesign.txt is provided here: https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed1.txt.

• First, empty rows and columns are removed from the data frame.

• Next, if a proteinGroups.txt file is used, it filters out reverse proteins, proteins that were only identified by site, and potential contaminants.Then it removes proteins identified with less than the number of unique peptides indicated by uniq_pep from the data frame.

• Next, it extracts the intensity columns indicated by data type and the selected protein rows from the data frame.

• Converts missing values (zeros) to NAs.

• Finally, the function log transforms the intensity values.

Value

A raw_df object which is a data frame containing protein intensities. Proteins or protein groups are indicated by rows and samples by columns.

Author(s)

Chathurani Ranathunge

Examples

### Using a proteinGroups.txt file produced by MaxQuant as input.
## Generate a raw_df object with default settings. No technical replicates.
raw_df <- create_df(
  prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg1.txt",
  exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed1.txt",
  input_type = "MaxQuant"
)

## Data containing technical replicates
raw_df <- create_df(
  prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg2.txt",
  exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed2.txt",
  input_type = "MaxQuant",
  tech_reps = TRUE
)

## Alter the number of unique peptides needed to retain a protein
raw_df <- create_df(
  prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg1.txt",
  exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed1.txt",
  input_type = "MaxQuant",
  uniq_pep = 1
)

## Use "iBAQ" values instead of "LFQ" values
raw_df <- create_df(
  prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg1.txt",
  exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed1.txt",
  input_type = "MaxQuant",
  data_type = "iBAQ"
)

### Using a universal standard input file instead of MaxQuant output.
raw_df <- create_df(
  prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/st.txt",
  exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed1.txt",
  input_type = "standard"
)

Cox et al 2014 LFQ data (fit object)

Description

An object of class "MArrayLM" from running find_dep on ecoli_norm_df

Usage

data(ecoli_fit_df)

Format

An object of class "MArrayLM"

References

https://europepmc.org/article/MED/24942700#id609082

Cox et al 2014 LFQ data (normalized)

Description

A dataframe containing normalized LFQ protein intensity data for 4360 proteins in 6 samples

Usage

data(ecoli_norm_df)

Format

A data frame with 4360 rows (proteins) and 6 columns (samples)

References

https://europepmc.org/article/MED/24942700#id609082

Visualize feature (protein) variation among conditions

Description

This function visualizes protein intensity differences among conditions (classes) using box plots or density distribution plots.

Usage

feature_plot(
  model_df,
  type = "box",
  text_size = 10,
  palette = "viridis",
  n_row,
  n_col,
  save = FALSE,
  file_path = NULL,
  file_name = "Feature_plot",
  file_type = "pdf",
  dpi = 80,
  plot_width = 7,
  plot_height = 7
)

Arguments

Details

This function visualizes condition-wise differences in protein intensity using boxplots and/or density plots.

Value

A ggplot2 object

Author(s)

Chathurani Ranathunge

See Also

• pre_process, rem_feature

Examples

## Create a model_df object with default settings.
covid_model_df <- pre_process(covid_fit_df, covid_norm_df)

## Feature variation - box plots
feature_plot(covid_model_df, type = "box", n_row = 4, n_col = 2)

## Density plots
feature_plot(covid_model_df, type = "density")

## Change color palette
feature_plot(covid_model_df, type = "density", n_row = 4, n_col = 2, palette = "rocket")

Filter proteins by group level missing data

Description

This function filters out proteins based on missing data at the group level.

Usage

filterbygroup_na(raw_df, set_na = 0.34, filter_condition = "either")

Arguments

Details

• This function first extracts group or condition information from the raw_df object and assigns samples to their groups.

• If filter_condition = "each", it then removes proteins (rows) from the data frame if the proportion of NAs in each group exceeds the threshold indicated by set_na (default is 0.34). This option is more lenient in comparison to filter_condition = "either", where proteins that exceeds the missing data threshold in either group gets removed from the data frame.

Value

A raw_df object.

Author(s)

Chathurani Ranathunge

See Also

create_df

Examples

# Generate a raw_df object with default settings. No technical replicates.
raw_df <- create_df(
prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg1.txt",
exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed1.txt"
)

## Remove proteins that exceed 34% NAs in either group (default)
rawdf_filt1 <- filterbygroup_na(raw_df)

## Remove proteins that exceed 34% NAs in each group
rawdf_filt2 <- filterbygroup_na(raw_df, filter_condition = "each")

## Proportion of samples with NAs allowed in each group = 0.5
rawdf_filt3 <- filterbygroup_na(raw_df, set_na = 0.5, filter_condition = "each")

Identify differentially expressed proteins between groups

Description

This function performs differential expression analysis on protein intensity data with limma.

Usage

find_dep(
  df,
  save_output = FALSE,
  save_tophits = FALSE,
  file_path = NULL,
  adj_method = "BH",
  cutoff = 0.05,
  lfc = 1,
  n_top = 20
)

Arguments

Details

• It is important that the data is first log-transformed, ideally, imputed, and normalized before performing differential expression analysis.

• save_output saves the complete results table from the differential expression analysis.

• save_tophits first subsets the results to those with absolute log fold change of more than 1, performs multiple correction with the method specified in adj_method and outputs the top n_top results based on lowest p-value and adjusted p-value.

• If the number of hits with absolute log fold change of more than 1 is less than n_top, find_dep prints only those with log-fold change > 1 to "TopHits.txt".

• If the file_path is not specified, text files will be saved in a temporary directory.

Value

A fit_df object, which is similar to a limma fit object.

Author(s)

Chathurani Ranathunge

References

Ritchie, Matthew E., et al. "limma powers differential expression analyses for RNA-sequencing and microarray studies." Nucleic acids research 43.7 (2015): e47-e47.

See Also

• normalize_data

• lmFit, eBayes, topTable, and write.fit functions from the limma package.

Examples

## Perform differential expression analysis using default settings
fit_df1 <- find_dep(ecoli_norm_df)

## Change p-value and adjusted p-value cutoff
fit_df2 <- find_dep(ecoli_norm_df, cutoff = 0.1)

Heatmap of differentially expressed proteins

Description

This function generates a heatmap to visualize differentially expressed proteins between groups

Usage

heatmap_de(
  fit_df,
  df,
  adj_method = "BH",
  cutoff = 0.05,
  lfc = 1,
  sig = "adjP",
  n_top = 20,
  palette = "viridis",
  text_size = 10,
  save = FALSE,
  file_path = NULL,
  file_name = "HeatmapDE",
  file_type = "pdf",
  dpi = 80,
  plot_height = 7,
  plot_width = 7
)

Arguments

Details

By default the tiles in the heatmap are reordered by intensity values along both axes (x axis = samples, y axis = proteins).

Value

A ggplot2 plot object.

Author(s)

Chathurani Ranathunge

See Also

• find_dep

• topTable and lmFit functions from the limma package.

Examples

## Build a heatmap of differentially expressed proteins using the provided
## example fit_df and norm_df data objects
heatmap_de(covid_fit_df, covid_norm_df)

## Create a heatmap with P-value of 0.05 and log fold change of 1 as
## significance criteria.
heatmap_de(covid_fit_df, covid_norm_df, cutoff = 0.05, sig = "P")

## Visualize the top 30 differentially expressed proteins in the heatmap and
## change the color palette
heatmap_de(covid_fit_df, covid_norm_df,
  cutoff = 0.05, sig = "P", n_top = 30,
  palette = "magma"
)

Visualize missing data

Description

This function visualizes the patterns of missing value occurrence using a heatmap.

Usage

heatmap_na(
  raw_df,
  protein_range,
  sample_range,
  reorder_x = FALSE,
  reorder_y = FALSE,
  x_fun = mean,
  y_fun = mean,
  palette = "viridis",
  label_proteins = FALSE,
  text_size = 10,
  save = FALSE,
  file_type = "pdf",
  file_path = NULL,
  file_name = "Missing_data_heatmap",
  plot_width = 15,
  plot_height = 15,
  dpi = 80
)

Arguments

Details

This function visualizes patterns of missing value occurrence using a heatmap. The user can choose to reorder the axes using the available functions (x_fun, y_fun) to better understand the underlying cause of missing data.

Value

A ggplot2 plot object.

Author(s)

Chathurani Ranathunge

See Also

create_df

Examples

## Generate a raw_df object with default settings. No technical replicates.
raw_df <- create_df(
  prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg1.txt",
  exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed1.txt"
)

## Missing data heatmap with default settings.
heatmap_na(raw_df)

## Missing data heatmap with x and y axes reordered by the mean (default) of
## protein intensity.
heatmap_na(raw_df,
  reorder_x = TRUE, reorder_y = TRUE
)

## Missing data heatmap with x and y axes reordered by the sum of
## protein intensity.
heatmap_na(raw_df,
  reorder_x = TRUE, reorder_y = TRUE, x_fun = sum,
  y_fun = sum
)

## Missing data heatmap for a subset of the proteins with x and y axes
## reordered by the mean (default) of protein intensity and the y axis
## labeled with protein IDs.
heatmap_na(raw_df,
  protein_range = 1:30,
  reorder_x = TRUE, reorder_y = TRUE,
  label_proteins = TRUE
)

Impute missing values

Description

This function imputes missing values using a user-specified imputation method.

Usage

impute_na(
  df,
  method = "minProb",
  tune_sigma = 1,
  q = 0.01,
  maxiter = 10,
  ntree = 20,
  n_pcs = 2,
  seed = NULL
)

Arguments

Details

• Ideally, you should first remove proteins with high levels of missing data using the filterbygroup_na function before running impute_na on the raw_df object or the norm_df object.

• impute_na function imputes missing values using a user-specified imputation method from the available options, minProb, minDet, kNN, RF, and SVD.

• Note: Some imputation methods may require that the data be normalized prior to imputation.

• Make sure to fix the random number seed with seed for reproducibility

.

Value

An imp_df object, which is a data frame of protein intensities with no missing values.

Author(s)

Chathurani Ranathunge

References

Lazar, Cosmin, et al. "Accounting for the multiple natures of missing values in label-free quantitative proteomics data sets to compare imputation strategies." Journal of proteome research 15.4 (2016): 1116-1125.

See Also

More information on the available imputation methods can be found in their respective packages.

• create_df

• For minProb and minDet methods, see imputeLCMD package.

• For Random Forest (RF) method, see missForest.

• For kNN method, see kNN from the VIM package.

• For SVD method, see pca from the pcaMethods package.

Examples

## Generate a raw_df object with default settings. No technical replicates.
raw_df <- create_df(
  prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg1.txt",
  exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed1.txt"
)

## Impute missing values in the data frame using the default minProb
## method.
imp_df1 <- impute_na(raw_df, seed = 3312)


## Impute using the RF method with the number of iterations set at 5
## and number of trees set at 100.
imp_df2 <- impute_na(raw_df,
  method = "RF",
  maxiter = 5, ntree = 100,
  seed = 3312
)


## Using the kNN method.
imp_df3 <- impute_na(raw_df, method = "kNN", seed = 3312)



## Using the SVD method with n_pcs set to 3.
imp_df4 <- impute_na(raw_df, method = "SVD", n_pcs = 3, seed = 3312)

## Using the minDet method with q set at 0.001.
imp_df5 <- impute_na(raw_df, method = "minDet", q = 0.001, seed = 3312)

## Impute a normalized data set using the kNN method
imp_df6 <- impute_na(ecoli_norm_df, method = "kNN")

Visualize the impact of imputation

Description

This function generates density plots to visualize the impact of missing data imputation on the data.

Usage

impute_plot(
  original,
  imputed,
  global = TRUE,
  text_size = 10,
  palette = "viridis",
  n_row,
  n_col,
  save = FALSE,
  file_path = NULL,
  file_name = "Impute_plot",
  file_type = "pdf",
  plot_width = 7,
  plot_height = 7,
  dpi = 80
)

Arguments

Details

• Given two data frames, one with missing values and the other, an imputed data frame (imp_df object) of the same data set, impute_plot generates global or sample-wise density plots to visualize the impact of imputation on the data set.

• Note, when sample-wise option is selected (global = FALSE), n_col and n_row can be used to specify the number of columns and rows to print the plots.

• If you choose to specify n_row and n_col, make sure that n_row * n_col matches the total number of samples in the data frame.

Value

A ggplot2 plot object.

Author(s)

Chathurani Ranathunge

Examples

## Generate a raw_df object with default settings. No technical replicates.
raw_df <- create_df(
  prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg1.txt",
  exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed1.txt"
)

## Impute missing values in the data frame using the default minProb
## method.
imp_df <- impute_na(raw_df)

## Visualize the impact of missing data imputation with a global density
## plot.
impute_plot(original = raw_df, imputed = imp_df)

## Make sample-wise density plots
impute_plot(raw_df, imp_df, global = FALSE)

## Print plots in user-specified numbers of rows and columns
impute_plot(raw_df, imp_df, global = FALSE, n_col = 2, n_row = 3)

Visualize the effect of normalization

Description

This function visualizes the impact of normalization on the data

Usage

norm_plot(
  original,
  normalized,
  type = "box",
  text_size = 10,
  palette = "viridis",
  save = FALSE,
  file_path = NULL,
  file_name = "Norm_plot",
  file_type = "pdf",
  dpi = 80,
  plot_width = 10,
  plot_height = 7
)

Arguments

Details

Given two data frames, one with data prior to normalization (original), and the other, after normalization (normalized), norm_plot generates side-by-side plots to visualize the effect of normalization on the protein intensity data.

Value

A ggplot2 plot object.

Author(s)

Chathurani Ranathunge

See Also

• normalize_data

• create_df

• impute_na

Examples

## Generate a raw_df object with default settings. No technical replicates.
raw_df <- create_df(
  prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg1.txt",
  exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed1.txt"
)

## Impute missing values in the data frame using the default minProb
## method.
imp_df <- impute_na(raw_df)

## Normalize the imp_df object using the default quantile method
norm_df <- normalize_data(imp_df)

## Visualize normalization using box plots
norm_plot(original = imp_df, normalized = norm_df)

## Visualize normalization using density plots
norm_plot(imp_df, norm_df, type = "density")

Normalize intensity data

Description

This function normalizes data using a user-specified normalization method.

Usage

normalize_data(df, method = "quantile")

Arguments

Details

• normalize_data is a wrapper function around the normalizeBetweenArrays function from the limma package.

• This function normalizes intensity values to achieve consistency among samples.

• It assumes that the intensities in the data frame have been log-transformed, therefore, it is important to make sure that create_df was run with log_tr = TRUE(default) when creating the raw_df object.

Value

A norm_df object, which is a data frame of normalized protein intensities.

Author(s)

Chathurani Ranathunge

See Also

• create_df

• impute_na

• See normalizeBetweenArrays in the R package limma for more information on the different normalization methods available.

Examples

## Generate a raw_df object with default settings. No technical replicates.
raw_df <- create_df(
  prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg1.txt",
  exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed1.txt"
)

## Impute missing values in the data frame using the default minProb
## method prioir to normalization.
imp_df <- impute_na(raw_df)

## Normalize the imp_df object using the default quantile method
norm_df1 <- normalize_data(imp_df)

## Use the cyclicloess method
norm_df2 <- normalize_data(imp_df, method = "cyclicloess")

## Normalize data in the raw_df object prior to imputation.
norm_df3 <- normalize_data(raw_df)

Proteins that are only expressed in a given group

Description

This function outputs a list of proteins that are only expressed (present) in one user-specified group while not expressed (completely absent) in another user-specified group.

Usage

onegroup_only(
  raw_df,
  abs_group,
  pres_group,
  set_na = 0.34,
  save = FALSE,
  file_path = NULL
)

Arguments

Details

Note: onegroup_only function assumes that column names in the raw_df object provided as df follow "Group_UniqueSampleID" notation. (Use head(raw_df) to check the structure of your raw_df object.)

• Given a pair of groups, onegroup_only function finds proteins that are only expressed in pres_group while completely absent or not expressed in abs_group.

• A text file containing majority protein IDs will be saved in a temporary directory if file_path is not specified.

Value

A list of majority protein IDs.

Author(s)

Chathurani Ranathunge

Examples

# Generate a raw_df object with default settings. No technical replicates.
raw_df <- create_df(
prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg1.txt",
exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed1.txt"
)

## Find the proteins only expressed in group L, but absent in group H.
onegroup_only(raw_df, abs_group = "H",
pres_group = "L")

Model performance plot

Description

This function generates plots to visualize model performance

Usage

performance_plot(
  model_list,
  type = "box",
  text_size = 10,
  palette = "viridis",
  save = FALSE,
  file_path = NULL,
  file_name = "Performance_plot",
  file_type = "pdf",
  plot_width = 7,
  plot_height = 7,
  dpi = 80
)

Arguments

Details

• performance_plot uses resampling results from models included in the model_list to generate plots showing model performance.

• The default metrics used for classification based models are "Accuracy" and "Kappa."

• These metric types can be changed by providing additional arguments to the train_models function. See train and trainControl for more information.

Value

A ggplot2 object.

Author(s)

Chathurani Ranathunge

See Also

• train_models

• resamples

• train

• trainControl

Examples

## Create a model_df object
covid_model_df <- pre_process(covid_fit_df, covid_norm_df)

## Split the data frame into training and test data sets
covid_split_df <- split_data(covid_model_df)

## Fit models based on the default list of machine learning (ML) algorithms
covid_model_list <- train_models(covid_split_df)

## Generate box plots to visualize performance of different ML algorithms
performance_plot(covid_model_list)

## Generate dot plots
performance_plot(covid_model_list, type = "dot")

## Change color palette
performance_plot(covid_model_list, type = "dot", palette = "inferno")

Pre-process protein intensity data for modeling

Description

This function pre-processes protein intensity data from the top differentially expressed proteins identified with find_dep for modeling.

Usage

pre_process(
  fit_df,
  norm_df,
  sig = "adjP",
  sig_cutoff = 0.05,
  fc = 1,
  n_top = 20,
  find_highcorr = TRUE,
  corr_cutoff = 0.9,
  save_corrmatrix = FALSE,
  file_path = NULL,
  rem_highcorr = TRUE
)

Arguments

Details

This function creates a data frame that contains protein intensities for a user-specified number of top differentially expressed proteins.

• Using find_highcorr = TRUE, highly correlated proteins can be identified, and can be removed with rem_highcorr = TRUE.

• Note: Most models will benefit from reducing correlation between proteins (predictors or features), therefore we recommend removing those proteins at this stage to reduce pairwise-correlation.

• If no or few proteins meet the significance threshold for differential expression, you may adjust sig, fc, and/or sig_cutoff accordingly to obtain more proteins for modeling.

Value

A model_df object, which is a data frame of protein intensities with proteins indicated by columns.

Author(s)

Chathurani Ranathunge

See Also

• find_dep, normalize_data

• caret: findCorrelation

Examples

## Create a model_df object with default settings.
covid_model_df1 <- pre_process(fit_df = covid_fit_df, norm_df = covid_norm_df)

## Change the correlation cutoff.
covid_model_df2 <- pre_process(covid_fit_df, covid_norm_df, corr_cutoff = 0.95)

## Change the significance criteria to include more proteins
covid_model_df3 <- pre_process(covid_fit_df, covid_norm_df, sig = "P")

## Change the number of top differentially expressed proteins to include
covid_model_df4 <- pre_process(covid_fit_df, covid_norm_df, sig = "P", n_top = 24)

Remove user-specified proteins (features) from a data frame

Description

This function removes user-specified proteins from a model_df object

Usage

rem_feature(model_df, rem_protein)

Arguments

Details

• After visualizing protein intensity variation among conditions with feature_plot or after assessing the importance of each protein in models using varimp_plot, you can choose to remove specific proteins (features) from the data frame.

• For example, you can choose to remove a protein from the model_df object if the protein does not show distinct patterns of variation among conditions. This protein may show mostly overlapping distributions in the feature plots.

• Another incidence would be removing a protein that is very low in variable importance in the models built using train_models. You can visualize variable importance using varimp_plot.

Value

A model_df object.

Author(s)

Chathurani Ranathunge

See Also

feature_plot, pre_process

Examples

covid_model_df <- pre_process(fit_df = covid_fit_df, norm_df = covid_norm_df)

## Remove sp|P22352|GPX3_HUMAN protein from the model_df object
covid_model_df1 <- rem_feature(covid_model_df, rem_protein = "sp|P22352|GPX3_HUMAN")

Remove user-specified samples

Description

This function removes user-specified samples from the data frame.

Usage

rem_sample(raw_df, rem)

Arguments

Details

• rem_sample assumes that sample names follow the "Group_UniqueSampleID_TechnicalReplicate" notation (Use head(raw_df) to see the structure of the raw_df object.)

• If all the technical replicates representing a sample needs to be removed, provide "Group_UniqueSampleID" as rem.

• If a specific technical replicate needs to be removed in case it shows weak correlation with other technical replicates for example, you can remove that particular technical replicate by providing "Group_UniqueSampleID_TechnicalReplicate" as rem.

Value

A raw_df object.

Author(s)

Chathurani Ranathunge

See Also

corr_plot, create_df

Examples

## Use a data set containing technical replicates to create a raw_df object
raw_df <- create_df(
prot_groups = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/pg2.txt",
exp_design = "https://raw.githubusercontent.com/caranathunge/promor_example_data/main/ed2.txt",
tech_reps = TRUE
)
# Check the first few rows of the raw_df object
head(raw_df)

## Remove all technical replicates of "WT_4"
raw_df1 <- rem_sample(raw_df, "WT_4")

## Remove only technical replicate number 2 of "WT_4"
raw_df2 <- rem_sample(raw_df, "WT_4_2")

ROC plot

Description

This function generates Receiver Operating Characteristic (ROC) curves to evaluate models

Usage

roc_plot(
  probability_list,
  split_df,
  ...,
  multiple_plots = TRUE,
  text_size = 10,
  palette = "viridis",
  save = FALSE,
  file_path = NULL,
  file_name = "ROC_plot",
  file_type = "pdf",
  plot_width = 7,
  plot_height = 7,
  dpi = 80
)

Arguments

Details

• roc_plot first uses probabilities generated during test_models to build a ROC object.

• Next, relevant information is extracted from the ROC object to plot the ROC curves.

Value

A ggplot2 object.

Author(s)

Chathurani Ranathunge

See Also

• test_models

• roc

Examples

## Create a model_df object
covid_model_df <- pre_process(covid_fit_df, covid_norm_df)

## Split the data frame into training and test data sets
covid_split_df <- split_data(covid_model_df)

## Fit models using the default list of machine learning (ML) algorithms
covid_model_list <- train_models(covid_split_df)

# Test a list of models on a test data set and output class probabilities,
covid_prob_list <- test_models(covid_model_list, covid_split_df, type = "prob")

## Plot ROC curves separately for each ML algorithm
roc_plot(covid_prob_list, covid_split_df)

## Plot all ROC curves in one plot
roc_plot(covid_prob_list, covid_split_df, multiple_plots = FALSE)

## Change color palette
roc_plot(covid_prob_list, covid_split_df, palette = "plasma")

Split the data frame to create training and test data

Description

This function can be used to create balanced splits of the protein intensity data in a model_df object to create training and test data

Usage

split_data(model_df, train_size = 0.8, seed = NULL)

Arguments

Details

This function splits the model_df object in to training and test data sets using random sampling while preserving the original class distribution of the data. Make sure to fix the random number seed with seed for reproducibility

Value

A list of data frames.

Author(s)

Chathurani Ranathunge

See Also

• pre_process

• createDataPartition

Examples

## Create a model_df object
covid_model_df <- pre_process(covid_fit_df, covid_norm_df)

## Split the data frame into training and test data sets using default settings
covid_split_df1 <- split_data(covid_model_df, seed = 8314)

## Split the data frame into training and test data sets with 70% of the
## data in training and 30% in test data sets
covid_split_df2 <- split_data(covid_model_df, train_size = 0.7, seed = 8314)

## Access training data set
covid_split_df1$training

## Access test data set
covid_split_df1$test

Test machine learning models on test data

Description

This function can be used to predict test data using models generated by different machine learning algorithms

Usage

test_models(
  model_list,
  split_df,
  type = "prob",
  save_confusionmatrix = FALSE,
  file_path = NULL,
  ...
)

Arguments

Details

• test_models function uses models obtained from train_models to predict a given test data set.

• Setting type = "raw" is required to obtain confusion matrices.

• Setting type = "prob" (default) will output a list of probabilities that can be used to generate ROC curves using roc_plot.

Value

• probability_list: If type = "prob", a list of data frames containing class probabilities for each method in the model_list will be returned.

• prediction_list: If type = "raw", a list of factors containing class predictions for each method will be returned.

Author(s)

Chathurani Ranathunge

See Also

• split_df

• train_models

• predict

• confusionMatrix

Examples

## Create a model_df object
covid_model_df <- pre_process(covid_fit_df, covid_norm_df)

## Split the data frame into training and test data sets
covid_split_df <- split_data(covid_model_df)

## Fit models using the default list of machine learning (ML) algorithms
covid_model_list <- train_models(covid_split_df)

# Test a list of models on a test data set and output class probabilities,
covid_prob_list <- test_models(model_list = covid_model_list, split_df = covid_split_df)


## Not run: 
# Save confusion matrices in the working directory and output class predictions
covid_pred_list <- test_models(
  model_list = covid_model_list,
  split_df = covid_split_df,
  type = "raw",
  save_confusionmatrix = TRUE,
  file_path = "."
)

## End(Not run)

Train machine learning models on training data

Description

This function can be used to train models on protein intensity data using different machine learning algorithms

Usage

train_models(
  split_df,
  resample_method = "repeatedcv",
  resample_iterations = 10,
  num_repeats = 3,
  algorithm_list,
  seed = NULL,
  ...
)

Arguments

Details

• train_models function can be used to first define the control parameters to be used in training models, calculate resampling-based performance measures for models based on a given set of machine-learning algorithms, and output the best model for each algorithm.

• In the event that algorithm_list is not provided, a default list of four classification-based machine-learning algorithms will be used for building and training models. Default algorithm_list: "svmRadial", "rf", "glm", "xgbLinear, and "naive_bayes."

• Note: Models that fail to build are removed from the output.

• Make sure to fix the random number seed with seed for reproducibility

Value

A list of class train for each machine-learning algorithm. See train for more information on accessing different elements of this list.

Author(s)

Chathurani Ranathunge

References

Kuhn, Max. "Building predictive models in R using the caret package." Journal of statistical software 28 (2008): 1-26.

See Also

• pre_process

• trainControl

• train

Examples

## Create a model_df object
covid_model_df <- pre_process(covid_fit_df, covid_norm_df)

## Split the data frame into training and test data sets
covid_split_df <- split_data(covid_model_df, seed = 8314)

## Fit models based on the default list of machine learning (ML) algorithms
covid_model_list1 <- train_models(split_df = covid_split_df, seed = 351)

## Fit models using a user-specified list of ML algorithms.
covid_model_list2 <- train_models(
  covid_split_df,
  algorithm_list = c("svmRadial", "glmboost"),
  seed = 351
)

## Change resampling method and resampling iterations.
covid_model_list3 <- train_models(
  covid_split_df,
  resample_method = "cv",
  resample_iterations = 50,
  seed = 351
)

Variable importance plot

Description

This function visualizes variable importance in models

Usage

varimp_plot(
  model_list,
  ...,
  type = "lollipop",
  text_size = 10,
  palette = "viridis",
  n_row,
  n_col,
  save = FALSE,
  file_path = NULL,
  file_name = "VarImp_plot",
  file_type = "pdf",
  dpi = 80,
  plot_width = 7,
  plot_height = 7
)

Arguments

Details

• varimp_plot produces a list of plots showing variable importance measures calculated from models generated with different machine-learning algorithms.

• Note: Variables are ordered by variable importance in descending order, and by default, importance values are scaled to 0 and 100. This can be changed by specifying scale = FALSE. See varImp for more information.

Value

A list of ggplot2 objects.

Author(s)

Chathurani Ranathunge

See Also

• train_models, rem_feature

• varImp

Examples

## Create a model_df object
covid_model_df <- pre_process(covid_fit_df, covid_norm_df)

## Split the data frame into training and test data sets
covid_split_df <- split_data(covid_model_df)

## Fit models based on the default list of machine learning (ML) algorithms
covid_model_list <- train_models(covid_split_df)

## Variable importance - lollipop plots
varimp_plot(covid_model_list)

## Bar plots
varimp_plot(covid_model_list, type = "bar")

## Do not scale variable importance values
varimp_plot(covid_model_list, scale = FALSE)

## Change color palette
varimp_plot(covid_model_list, palette = "magma")

Volcano plot

Description

This function generates volcano plots to visualize differentially expressed proteins between groups.

Usage

volcano_plot(
  fit_df,
  adj_method = "BH",
  sig = "adjP",
  cutoff = 0.05,
  lfc = 1,
  line_fc = TRUE,
  line_p = TRUE,
  palette = "viridis",
  text_size = 10,
  label_top = FALSE,
  n_top = 10,
  save = FALSE,
  file_path = NULL,
  file_name = "Volcano_plot",
  file_type = "pdf",
  plot_height = 7,
  plot_width = 7,
  dpi = 80
)

Arguments

Details

• Volcano plots show log-2-fold change on the x-axis, and based on the significance criteria chosen, either -log10(p-value) or -log10(adjusted p-value) on the y-axis.

• volcano_plot requires a fit_df object from performing differential expression analysis with find_dep.

• User has the option to choose criteria that denote significance.

Value

A ggplot2 plot object.

Author(s)

Chathurani Ranathunge

See Also

• find_dep

• topTable and lmFit functions from the limma package.

Examples

## Create a volcano plot with default settings.
volcano_plot(ecoli_fit_df)

## Change significance criteria and cutoff
volcano_plot(ecoli_fit_df, cutoff = 0.1, sig = "P")

## Label top 30 differentially expressed proteins and
## change the color palette of the plot
volcano_plot(ecoli_fit_df, label_top = TRUE, n_top = 30, palette = "mako")