Normalize Phyloseq Data

This function applies normalization to a phyloseq object, converting data to absolute values based on 16S rRNA copy numbers and sample biomass, using either flow cytometry (FCM) data or qPCR data. The function can apply copy number correction prior to biomass normalization.

Usage

normalise_data(
  physeq = without_mock_physeq,
  norm_method = NULL,
  copy_correction = TRUE
)

Arguments

physeq

A phyloseq object containing the microbiome data. This is the input object that the function processes.

norm_method

A character string specifying the normalization method. Options are:

• "fcm": Normalize based on flow cytometry data, converting abundances to cell concentrations (cells/mL or per gram sample).

• "qpcr": Normalize based on qPCR data, converting abundances to cell equivalents (cells/mL or per gram sample).

• NULL: Apply only copy number correction without further normalization.

copy_correction

A logical value indicating whether the data should be corrected for the predicted 16S rRNA copy numbers prior to biomass normalization. Options are:

• TRUE: Both relative and absolute abundances are corrected using the predicted copy numbers.

• FALSE: Abundances are not corrected by copy number. Note that qPCR normalization requires copy number correction to provide absolute data.

Value

The function saves multiple phyloseq objects as RDS files:

• <project_name>_phyloseq_asv_level_without_copy_number_corrected_counts.rds: if copy_correction = FALSE; a phyloseq object with uncorrected counts.

• <project_name>_phyloseq_asv_level_copy_number_corrected_counts.rds: if copy_correction = TRUE; a phyloseq object with counts corrected by predicted 16S copy numbers.

• <project_name>_phyloseq_asv_level_fcm_normalised_cell_concentration.rds: if norm_method = "fcm" and copy_correction = TRUE; a phyloseq object with abundances normalized based on FCM data and copy number correction.

• <project_name>_phyloseq_asv_level_fcm_normalised_cell_concentration_without_copy_number_corrected_count.rds: if norm_method = "fcm" and copy_correction = FALSE; a phyloseq object with FCM normalization applied without copy number correction.

• <project_name>_phyloseq_asv_level_qpcr_normalised_cell_concentration.rds: if norm_method = "qpcr"; a phyloseq object with abundances normalized to cell equivalents using qPCR data.

The relative phyloseq object (without biomass normalization) is saved in the output_data/rds_files/After_cleaning_rds_files/ASV directory, and all biomass-normalized phyloseq objects are saved in the output_data/rds_files/Before_cleaning_rds_files directory.

Details

The function follows these steps based on the chosen parameters:

1. Copy Number Correction (if copy_correction = TRUE):

   • Correct ASV abundances by dividing each count by its predicted 16S rRNA copy number. The prediction is based on the method described in "Accounting for 16S rRNA copy number prediction uncertainty and its implications in bacterial diversity analyses".

   • This correction adjusts for variability in 16S rRNA gene copy numbers across taxa, enabling the calculation of cell equivalents.

2. FCM Normalization (norm_method = "fcm"):

   • When copy_correction = TRUE: The copy number–corrected abundances are multiplied by the FCM data, where FCM data (cells per mL or per gram) is included in the metadata or a file with "fcm" in the name, with the column cells_per_ml.

   • When copy_correction = FALSE: The raw abundances are multiplied by the FCM data without prior copy number correction.

3. qPCR Normalization (norm_method = "qpcr"):

   • The qPCR data, provided in 16S copies per mL or per gram sample (included in the metadata or a file with "qpcr" in the name and column sq_calc_mean), is used together with copy number predictions to calculate absolute abundances.

DNA vs. RNA Normalization

The interpretation of normalized data depends on the nucleic acid type:

• DNA: Normalized abundances usually represent cells per mL (or per gram), assuming one genome copy per cell.

• RNA: Normalized abundances often represent copies per cell equivalent per mL (or per gram); RNA reflects transcriptional activity and may vary considerably with cell condition.

References

Gao, Y., & Wu, M. (2023). Accounting for 16S rRNA copy number prediction uncertainty and its implications in bacterial diversity analyses. ISME Communications, 3(1), 59. doi:10.1038/s43705-023-00266-0

Examples

if (FALSE) { # \dontrun{
# Apply only copy number correction
result <- normalize_data(physeq = physeq, norm_method = NULL)

# Normalize using flow cytometry (FCM) data
result <- normalize_data(physeq = physeq, norm_method = "fcm", copy_correction = TRUE)

# Normalize using qPCR data
result <- normalize_data(physeq = physeq, norm_method = "qpcr", copy_correction = TURE)
} # }