In alessandrozito/BayesANT: Bayesian Nonparametric Taxonomic classifier

knitr::opts_chunk$set(echo = TRUE)
library(BayesANT)
library(seqinr)

# Load aminoacid sequences
fasta.file <- "~/Downloads/transfer_284506_files_d322bf12/train_finbol-gbol_genus.translated.fasta"
df_train <- read.fasta(fasta.file)

df_train <- read.BayesANT.data(fasta.file = fasta.file, sep = '\\s+|\\|')
head(df_train)

# Train model
model <- BayesANT_amino(df_train, amino = TRUE)

# Predict
fasta.file_test <- "~/Downloads/transfer_284506_files_d322bf12/testshort_finbol-gbol_genus.translated.fasta"
df_test <- read.BayesANT.data(fasta.file = fasta.file_test, sep = '\\s+|\\|')
df_test <- add_novelty_to_test_data(df_train, df_test)
testSeqs <- read.BayesANT.testDNA(fasta.file_test)

nseqs <- 500
pred <- predict(model, DNA = testSeqs[1:nseqs], rho = 0.25)

plot_accuracies(prediction = pred, test_data = df_test[1:nseqs, ])

alessandrozito/BayesANT documentation built on April 5, 2025, 6:22 a.m.