gRodon

https://jlw-ecoevo.github.io/gRodon-vignette
Admin
The gRodon package predicts the maximal growth rate of prokaryotic organisms using codon usage...
Tokens:3,503
Snippets:36
Trust Score:9.7
Update:3 months ago
Show doc for...
Tokens:
Raw
### Install gRodon from GitHub

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Install the gRodon package directly from its GitHub repository using the devtools package.

```r
devtools::install_github("jlw-ecoevo/gRodon2")
```

--------------------------------

### Load Genome Assembly

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Load a genome assembly FASTA file using Biostrings. The example uses a file provided with the gRodon package.

```r
path_to_genome <- system.file('extdata',
  'GCF_000349925.2_ASM34992v2_cds_from_genomic.fna.gz',
  package = 'gRodon')
genes <- readDNAStringSet(path_to_genome)
```

--------------------------------

### Predict Growth with Temperature Correction

Source: https://jlw-ecoevo.github.io/gRodon-vignette

This example demonstrates predicting growth rates while accounting for the organism's optimal growth temperature. It's particularly useful for psychrophiles and thermophiles, though the correction is based on a limited dataset.

```R
predictGrowth(genes, highly_expressed, temperature = 33)
```

--------------------------------

### Install gRodon Dependencies

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Install necessary Bioconductor and CRAN dependencies for gRodon, including Biostrings, coRdon, and matrixStats. This ensures all required packages are available for gRodon's functionality.

```r
if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install("Biostrings")
BiocManager::install("coRdon")
install.packages("matrixStats")
```

--------------------------------

### Run Prokka for Genome Annotation

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Use prokka on the command line to annotate a genome assembly. This generates various output files, including CDS information.

```bash
prokka contig.fasta --prefix my_genome
```

--------------------------------

### Predict Growth with Madin Training Set

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Use the 'madin' training set for growth rate prediction, which is derived from minimal doubling time estimates compiled by Madin et al. This is now the default training set in recent gRodon updates. Ensure necessary packages are loaded and data is prepared.

```R
path_to_genome <- system.file('extdata',
  'GCF_000349925.2_ASM34992v2_cds_from_genomic.fna.gz',
  package = 'gRodon')
genes <- readDNAStringSet(path_to_genome)
highly_expressed <- grepl("ribosomal protein",names(genes),ignore.case = T)
predictGrowth(genes, highly_expressed, training_set = "madin")
#> Warning in filterSeq(genes = genes, highly_expressed = highly_expressed, : 
#> There were 146 genes either with lengths not multiples of 3 or not above length
#> threshold (default 240bp), these genes have been ignored
#> $CUBHE
#> [1] 0.9148924
#> 
#> $GC
#> [1] 0.3917167
#> 
#> $GCdiv
#> [1] 0.1082833
#> 
#> $ConsistencyHE
#> [1] 0.5062119
#> 
#> $CUB
#> [1] 0.564834
#> 
#> $CPB
#> [1] -0.3886505
#> 
#> $FilteredSequences
#> [1] 146
#> 
#> $nHE
#> [1] 45
#> 
#> $dCUB
#> [1] -0.6197544
#> 
#> $d
#> [1] 0.364274
#> 
#> $LowerCI
#> [1] 0.2976978
#> 
#> $UpperCI
#> [1] 0.4483901

```

--------------------------------

### Load gRodon and Biostrings Packages

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Load the necessary R packages for gRodon functionality. Biostrings is required for reading DNA sequences.

```r
library(gRodon)
library(Biostrings)
```

--------------------------------

### Predict Growth Rate with Weighted Metagenome Mode v2

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Estimate the average community-wide maximal growth rate using metagenome_v2 mode, incorporating relative organism abundance via the depth_of_coverage parameter for more realistic estimates.

```R
predictGrowth(genes, 
              highly_expressed, 
              mode = "metagenome_v2", 
              depth_of_coverage = depth_of_coverage)
```

--------------------------------

### Predict Growth with AOA_NOB Training Set

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Utilize the 'AOA_NOB' training set, which includes augmented data for Archaea (AOA) and Ammonia-Oxidizing Bacteria (NOB) from enrichment cultures. This set enhances gRodon's predictive capabilities for these specific microbial groups. Ensure data preparation is complete before calling the function.

```R
path_to_genome <- system.file('extdata',
  'GCF_000349925.2_ASM34992v2_cds_from_genomic.fna.gz',
  package = 'gRodon')
genes <- readDNAStringSet(path_to_genome)
highly_expressed <- grepl("ribosomal protein",names(genes),ignore.case = T)
predictGrowth(genes, highly_expressed, training_set = "AOA_NOB")
#> Warning in filterSeq(genes = genes, highly_expressed = highly_expressed, : 
#> There were 146 genes either with lengths not multiples of 3 or not above length
#> threshold (default 240bp), these genes have been ignored
#> $CUBHE
#> [1] 0.9148924
#> 
#> $GC
#> [1] 0.3917167
#> 
#> $GCdiv
#> [1] 0.1082833
#> 
#> $ConsistencyHE
#> [1] 0.5062119
#> 
#> $CUB
#> [1] 0.564834
#> 
#> $CPB
#> [1] -0.3886505
#> 
#> $FilteredSequences
#> [1] 146
#> 
#> $nHE
#> [1] 45
#> 
#> $dCUB
#> [1] -0.6197544
#> 
#> $d
#> [1] 0.3582411
#> 
#> $LowerCI
#> [1] 0.2937023
#> 
#> $UpperCI
#> [1] 0.4394778

```

--------------------------------

### Predict Growth Rate in Metagenome Mode v2

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Use this snippet to estimate the average community-wide maximal growth rate from metagenomic samples using the metagenome_v2 mode. Ensure you are using this mode for metagenomic data to avoid underestimation.

```R
path_to_metagenome <- system.file('extdata',
  'ERR2143764_fastp_prokka_scaffolds.ffn.gz',
  package = 'gRodon')
genes <- readDNAStringSet(path_to_metagenome)
highly_expressed <- grepl("ribosomal protein",names(genes),ignore.case = T)
predictGrowth(genes, highly_expressed, mode = "metagenome_v2")
```

--------------------------------

### Load Read Depths for Weighted Metagenome Mode

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Load and prepare read depth data from a TSV file. This data is used to weight the metagenome mode predictions by the relative abundance of different organisms.

```R
path_to_coverage <- system.file('extdata',
  'ERR2143764_fastp_map2ffn_counts.tsv',
  package = 'gRodon')
read_depths <- read.delim(path_to_coverage,
                            stringsAsFactors = FALSE)
depths <- read_depths$meandepth
#Make sure in the correct order
names(depths) <- read_depths$X.rname
depth_of_coverage <- depths[gsub(" .+", "", names(genes))]
```

--------------------------------

### Predict Maximal Growth Rate in Partial Mode

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Predict maximal growth rate using 'partial' mode, which excludes codon pair bias. This is recommended for incomplete genomes like SAGs and MAGs.

```r
predictGrowth(genes, highly_expressed, mode="partial")
```

--------------------------------

### Predict Growth for Slow Growers

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Use this snippet to predict growth rates from genes, especially when dealing with potentially slow-growing organisms. The function may issue a warning if the estimated doubling time exceeds 5 hours, indicating a potential underestimate.

```R
path_to_genome <- system.file('extdata',
  'GCF_003253775.1_ASM325377v1_cds_from_genomic.fna.gz',
  package = 'gRodon')
genes <- readDNAStringSet(path_to_genome)
highly_expressed <- grepl("ribosomal protein",names(genes),ignore.case = T)
predictGrowth(genes, highly_expressed)
```

--------------------------------

### Extract CDS IDs from GFF File

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Process the prokka GFF output to extract a list of CDS IDs. This command filters for CDS entries and outputs their IDs to a text file.

```bash
sed -n '/##FASTA/q;p' my_genome/my_genome.gff | awk '$3=="CDS"' | awk '{print $9}' | awk 'gsub(";.*","")' | awk 'gsub("ID=","")' > CDS_names.txt
```

--------------------------------

### Predict Maximal Growth Rate

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Predict the maximal growth rate using the loaded genes and the identified highly expressed genes. A warning indicates genes filtered due to length or codon count.

```r
predictGrowth(genes, highly_expressed)
```

--------------------------------

### Load and Filter Gene Sequences in R

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Load nucleotide sequences and CDS IDs into R. This snippet subsets the gene sequences to include only those identified as CDS and then identifies highly expressed genes, specifically looking for ribosomal proteins.

```r
# Load your *.ffn file into R
genes <- readDNAStringSet("my_genome/my_genome.ffn")

# Subset your sequences to those that code for proteins
CDS_IDs <- readLines("CDS_names.txt")
gene_IDs <- gsub(" .*","",names(genes)) #Just look at first part of name before the space
genes <- genes[gene_IDs %in% CDS_IDs]

#Search for genes annotated as ribosomal proteins
highly_expressed <- grepl("^(?!.*(methyl|hydroxy)).*0S ribosomal protein",names(genes),ignore.case = T, perl = TRUE)
```

--------------------------------

### Identify Highly Expressed Genes

Source: https://jlw-ecoevo.github.io/gRodon-vignette

Identify highly expressed genes by searching for 'ribosomal protein' in gene names. This uses a case-insensitive grepl function to return a logical vector.

```r
highly_expressed <- grepl("ribosomal protein",names(genes),ignore.case = T)
```
Tokens:
Raw