pah_pca: pah_pca
In limnoliver/pah: Common Analyses for Polycyclic Aromatic Hydrocarbons (PAH) Data

Conducts principle components analysis on the source and sample PAH profiles. This reduces the profiles to manageable dimensions. All important PCA axes are then used to calculate the Euclidean distance between source and samples to determine likely sources.

1	pah_pca(profiles, perc_cutoff = 10, sample_column)

`profiles`	The output of pah_profiles, which contains PAH profile data for both source and samples. The reported values are proportional concentrations for each compound.
`perc_cutoff`	The threshold for keeping PCA components in subsequent analyses; all individual components that explain greater than or equal to this cutoff of the total variance in the dataset will be kept in the analysis.
`sample_column`	column name of unique sample identifier

A list of three data frames. "pca_dat" contains all PCA components that met the perc_cutoff, as well as a sample_id column and type which identifies whether the values correspond to a sample or source profile. "pca_summary" prints the standard deviation, proportion of variance, and cumulative proportion of variance explained for all components of the PCA. "pca_distance" is a long data frame of all source-sample compindations, and the euclidean distance between the source and sample in the PCA components space.

limnoliver/pah documentation built on April 30, 2020, 2:45 p.m.