Data Preparation Modules: panoply_normalize_ms_data

panoply_normalize_ms_data

Description

This module normalizes proteomics (protein/PTM site) data. The normalization methods, especially two-component normalization, assume that the input data are log ratios to a common reference.

Normalization methods available are:

• Z-scoring (mean): mean-centering followed by standard deviation scaling
• Median-MAD normalization (median): median-centering followed by median absolute deviation (MAD) scaling
• Two-component mixture model-based normalization (2comp): In this method, we assume that for every sample there is a set of unregulated proteins or PTM sites. In the normalized sample, these proteins or PTM sites should have a log ratio centered at zero. In addition, there are proteins or PTM sites that are either up- or downregulated. This normalization scheme attempts to identify the unregulated proteins or PTM sites, and centers the distribution of these log-ratios around zero in order to nullify the effect of differential protein loading and/or systematic MS variation. A 2-component Gaussian mixture model-based is used to achieve this effect. The two Gaussians N (μ₁, σ₁) and N (μ₂, σ₂) for a sample i are fitted and used in the normalization process as follows: the mode m_i of the log-ratio distribution is determined for each sample using kernel density estimation with a Gaussian kernel and Shafer-Jones bandwidth. A two-component Gaussian mixture model is then fit with the mean of both Gaussians constrained to be m_i, i.e., μ_i1 = μ_i2 = m_i. The Gaussian with the smaller estimated standard deviation σ_i = min (σ_i1 , σ_i2) is assumed to represent the unregulated component of proteins/PTM sites, and is used to normalize the sample by subtracting the mean m_i from each protein/PTM site and dividing by the standard deviation σ_i. See (Mertins et al., 2016) and (Gillette et al, 2020).

CAVEAT: The two-component mixture model-based normalization (2comp) method has been tuned for log-transformed ratio (to a common reference) data, and hence cannot be used with intensity-based (e.g., label-free) data.

The module also creates

• Profile plots (density plots) for each sample before and after normalization for comparison.
• Plots of summary of normalizaton statistics.

Input

Required inputs:

• inputData: (.tar file) tarball from panoply_parse_sm_table or un-normalized input data in gct format (when standalone is true)
• type: (String) proteomics data type
• standalone: (String) set to true to run as a self-contained module; if true the analysisDir input is required
• yaml: (.yaml file) parameters in yaml format
• analysisDir: (String) name of analysis directory

Optional inputs:

• normalizeProteomics: (String, default chosen in startup notebook) when 'true' normalization will be applied, when 'false' normalization is skipped
• normMethod: (String, default = '2comp') normalization method; options are '2comp', 'median', 'mean'
• altMethod: (String, default = 'median') alternate normalization method for comparison with normMethod; downstream modules typically do not generate analyses for the data normalized using altMethod
• ndigits: (Int, default = 5) number of decimal digits to use in output tables
• outTar: (String, default = "panoply_normalize_ms_data-output.tar") output .tar file name
• outTable: (String, default = "normalized_table-output.gct") output .gct normalized file name

Output

• output_tar: Tarball including the following files in the normalized-data subdirectory:
  • Normalized data files:
    • normalized data table (*-ratio-norm.gct)
    • normalized data table using alternate normalization method specified in altMethod (*-ratio-*-norm.gct)
  • Plots and normalization statistics
    • profile plot (density of log ratio values for each sample) showing distribution for all samples in input data, before normalization (*-ratio-profile-plot.pdf)
    • profile plot showing distribution for samples after normalization, using primary normalization method (*-ratio-norm-profile-plot.pdf) and alternate normalization method (*-ratio-median-norm-profile-plot.pdf)
    • normalization statistics table showing centering and scaling factors for each sample, using primary normalization method (*-ratio-norm-stats.csv) and alternate normalization method (*-ratio-median-norm-stats.csv)
    • boxplot of normalization statistics for QC.pass and QC.fail samples (*-ratio-norm-stats.pdf and *-ratio-median-norm-stats.pdf)
• outputs: (.gct file) fnormalized data table (equivalent to *-ratio-norm.gct)
• output_yaml: finalized parameter file

References

• Mertins, P., Mani, D., Ruggles, K., Gillette, M., Clauser, K., Wang, P., Wang, X., Qiao, J., Cao, S., Petralia, F., et al. (2016). Proteogenomics connects somatic mutations to signalling in breast cancer. Nature 534(7605), 55 - 62. https://dx.doi.org/10.1038/nature18003.
• Gillette, M., Satpathy, S., Cao, S., Dhanasekaran, S., Vasaikar, S., Krug, K., Petralia, F., Li, Y., Liang, W., Reva, B., et al. (2020). Proteogenomic Characterization Reveals Therapeutic Vulnerabilities in Lung Adenocarcinoma. Cell 182(1), 200 - 225.e35. https://dx.doi.org/10.1016/j.cell.2020.06.013