SpectroPipeR: normalization & quantification module
norm_quant_module.RdFunction for normalizing and quantifying Spectronaut output reports, serving as the second step in the pipeline and building upon the initial step of reading Spectronaut data.
Usage
norm_quant_module(
SpectroPipeR_data = NULL,
batch_adjusting = FALSE,
sample__batch_meta_data_file = NULL,
batch_adjusting_column = "",
number_of_cores_adjusting = parallel::detectCores() - 2,
covariate_adjusting_formula = "",
covariate_adjusting_meta_data_file = "",
skipping_MaxLFQ_median_norm = FALSE,
costum_colors = NULL,
print.plot = FALSE
)Arguments
- SpectroPipeR_data
it is the SpectroPipeR_data list object generated by the read_spectronaut_module() function
- batch_adjusting
logical - if batch adjusting with ComBat (sva package) should be performed; default = FALSE
- sample__batch_meta_data_file
character - sample batch file; tab-delimited txt-file, containing "R.FileName" column e.g. sample__batch_meta_data_file = "Sample_MetaData_Batches.txt"
example table for batch meta data:
R.FileName digest_batch 20230403_TIMSTOF_1_S1-B11_1_6690 1 20230403_TIMSTOF_2_S1-G11_1_6695 1 20230403_TIMSTOF_3_S1-E7_1_6661 1 20230403_TIMSTOF_4_S1-A9_1_6673 2 20230403_TIMSTOF_7_S1-D8_1_6668 2 20230403_TIMSTOF_9_S1-D3_1_6627 2 A good starting point for the generation of the table is the '*_ConditionSetup.tsv' in your Spectronaut Pipeline Report export folder
- batch_adjusting_column
character - column name in sample__batch_meta_data_file, which should be used for assigning the samples to batches
- number_of_cores_adjusting
numeric - number of processor cores used for batch or covariate adjustment
- covariate_adjusting_formula
character - provide a formula passed to lm() for covariate adjustment e.g. "log10_peptide_intensity ~ log10(CRP)+log10(age)+as.factor(sex)"; you may also use ns() function e.g. "log10_peptide_intensity ~ ns(age, df=3)"
- covariate_adjusting_meta_data_file
covariate meta csv file, containing "R.FileName; age; sex;..."; you may find a start file in the 02_ID_rate folder > file_list.csv column e.g. covariate_adjusting_meta_data_file = "covariate_MetaData_file.csv"
example table for covariate meta data:
R.FileName R.FileName_raw R.Condition sex CRP 20230403_TIMS..._S1-B11_1_6690 20230403_TIMSTOF_1_S1-B11_1_6690 heathy 1 3.8 20230403_TIMS..._S1-G11_1_6695 20230403_TIMSTOF_2_S1-G11_1_6695 heathy 2 5.1 20230403_TIMS...3_S1-E7_1_6661 20230403_TIMSTOF_3_S1-E7_1_6661 heathy 1 1.2 20230403_TIMS...4_S1-A9_1_6673 20230403_TIMSTOF_4_S1-A9_1_6673 cancer 1 50.2 20230403_TIMS...7_S1-D8_1_6668 20230403_TIMSTOF_7_S1-D8_1_6668 cancer 2 30.8 20230403_TIMS...9_S1-D3_1_6627 20230403_TIMSTOF_9_S1-D3_1_6627 cancer 2 64.1 - skipping_MaxLFQ_median_norm
logical - if median normalization after MaxLFQ calculation should be skipped; default = FALSE; applied only if MaxLFQ protein estimation is selected
- costum_colors
if you would like to use your own colors please provide a named color vector (e.g. c(condition1 = "black", condition2 = "grey")); names should have the same naming and length like the conditions set in Spectronaut
- print.plot
logical - should a plot of normalization factors be printed
Value
SpectroPipeR_norm_quant list object with the loaded raw data and processed data tables, in addition to the automatically saved tables and plots For the description of the generated figures and tables please read the manual & vignettes
| list element | description |
| data_input_normalized | tibble: Spectronaut report tibble provided for the analysis |
| MedianNormalizationFactor | tibble: ion normalization factor table |
| MedianNormalizationFactor_outlier | tibble: table containing detected norm. outliers on ion level |
| NormFactor_plot | ggplot2 plot: norm. factor plots |
| iBAQ_intensities | tibble: table containing the iBAQ int. |
| iBAQ_intensities_summary | tibble: table containing the per condition summarized iBAQ int. |
| protein_data | tibble: protein intensity table (e.g. Hi3 or MaxLFQ int.) |
| PG_2_peptides_ID_raw | tibble: with protein groups with at least 2 peptides with peptide |
| and replicate count | |
| protein_data_normalization_factor | tibble: normalization factor table for protein int. data |
| peptide_intensity_filtered_2pep_hi3 | tibble: if Hi3 protein int. was selected a table containing |
| the peptides and intensities used for Hi3 protein intensity calculation | |
| peptide_intensity | tibble: peptides intensity table based on norm. ion intensity |
| parameter | list: parameters provided by the user updated with the |
| norm_quant_module() parameters | |
| CV_cumulative_frequency | tibble: cumulative frequency table on peptide |
| and protein intensity level | |
| sample_length | numberical value: number of samples in the provided Spectronaut report |
Details
batch adjustment
Batch effects refer to systematic differences between batches (groups) of samples in high-throughput experiments. These differences can arise due to various factors, such as batch variations in sample preparation, handling, processing procedures and measurement orders. Batch effects can obscure the true biological signal and lead to incorrect conclusions if not properly accounted for. In the SpectroPipeR pipeline, the ComBat tool was employed to adjust for batch effects in the datasets where the batch covariate was known. ComBat utilizes the methodology described in Johnson et al. 2007. It uses an empirical Bayes (EB) framework for adjusting data for batch effects that is robust to outliers in small sample sizes and performs comparable to existing methods for large samples. Johnson et al. 2007: This method incorporates systematic batch biases common across genes in making adjustments, assuming that phenomena resulting in batch effects often affect many genes in similar ways (i.e. increased expression, higher variability, etc). Specifically, the the L/S model parameters are estimated that represent the batch effects by pooling information across peptides in each batch to shrink the batch effect parameter estimates toward the overall mean of the batch effect estimates (across genes). These EB estimates are then used to adjust the data for batch effects, providing more robust adjustments for the batch effect on each peptide. In SpectroPipeR a parametric ComBAT emperical Bayes adjustment is implemented by utilizing the sva-package.
covariate adjustment
If a covariate adjustment of peptide intensity data was performed using the users input formula, a linear mixed model (LMM) was calculated based on that formula per peptide and the outcoming residuals were added to the mean peptide intensity over the samples. This means that the adjusted peptide intensities retain their intensity level (low intense peptides keep their low intensity and high intense ions keep their higher intensity).
Examples
# \donttest{
#load library
library(SpectroPipeR)
# use default parameters list
params <- list(output_folder = "../SpectroPipeR_test_folder")
# example input file
example_file_path <- system.file("extdata",
"SN_test_HYE_mix_file.tsv",
package="SpectroPipeR")
# step 1: load Spectronaut data module
SpectroPipeR_data <- read_spectronaut_module(file = example_file_path,
parameter = params,
print.plot = FALSE)
# step 2: normalize & quantification module
SpectroPipeR_data_quant <- norm_quant_module(SpectroPipeR_data = SpectroPipeR_data)
# }