predictGrowth: Predict Growth Rate

In jlw-ecoevo/gRodon: Inference of prokaryotic growth rates with codon usage statistics

Predict Growth Rate

Description

This function predicts the growth rate of a prokaryotic organism based on genomic codon usage patterns (Weissman et. al. TBD).

Usage

predictGrowth(
  genes,
  highly_expressed,
  mode = "full",
  temperature = "none",
  training_set = "vs",
  depth_of_coverage = NULL,
  fragments = FALSE
)

Arguments

genes	DNAStringSet object holding all nucleotide sequences from a genome. See Biostrings package.
highly_expressed	Logical vector describing the set of highly expressed genes. Must be of same length as genes.
mode	Whether to run prediction in full, partial, or metagenome mode (by default gRodon applies the full model)
temperature	Optimal growth temperature. By default this is set as "none" and we do not guarantee good results for non-mesophilic organisms since few were used to fit the model.
training_set	Whether to use models trained on the original Vieira-Silva et al. doubling time dataset or doubling times drawn from the Madin et al. database.
depth_of_coverage	When using metagenome mode, provide a vector containing the coverage of your ORFs to improve your estimate
fragments	If using gene fragments predicted from reads, will use a more permissive length filter (120bp as opposed to 240bp)

Value

gRodon returns a list with the following elements:

CUBHE

Median codon usage bias of the highly expressed genes (MILC) calculated using the genome-wide codon usage as the expected bias

ConsistencyHE

Mean codon usage bias of the highly expressed genes (MILC) calculated using the codon usage of highly expressed genes as the expected bias

CPB

Genome-wide codon pair bias (Coleman et al. 2008)

FilteredSequences

Number of gene sequences filtered out during calulation (due to length and/or presence of ambiguous bases)

d

Predicted doubling time in hours

LowerCI

Lower CI of d (2.5%)

UpperCI

Upper CI of d (97.5%)

Examples

# Load in example genome (Streptococcus pyogenes M1, downloaded from RefSeq)
# included with gRodon
path_to_genome <- system.file('extdata',
  'GCF_000349925.2_ASM34992v2_cds_from_genomic.fna.gz',
  package = 'gRodon')
genes <- readDNAStringSet(path_to_genome)

# Search pre-existing annotations for ribosomal proteins, which we
# will use as our set of highly expressed genes
highly_expressed <- grepl("ribosomal protein",names(genes),ignore.case = T)

# Run the gRodon growth prediction pipeline
predictGrowth(genes, highly_expressed)

# Run gRodon with temperature option (not needed for mesophiles, gRodon not
# validated on extremophiles, use with care)
predictGrowth(genes, highly_expressed, temperature = 37)

jlw-ecoevo/gRodon documentation built on Dec. 3, 2022, 11:39 a.m.