tests/testthat/test.annots.r

In surh/HMVAR: Human Microbiome Variant Analysis in R

# (C) Copyright 2019 Sur Herrera Paredes
# 
# This file is part of HMVAR.
# 
# HMVAR is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
# 
# HMVAR is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
# 
# You should have received a copy of the GNU General Public License
# along with HMVAR.  If not, see <http://www.gnu.org/licenses/>.

context("Everything about annotations")
library(HMVAR)

test_that('annots_to_geneGO',{
  d <- tibble::tibble(gene_id = c('gene1', 'gene2', 'gene3'),
                      terms = c(NA, 'term1,term2', 'term2, term3'))
  e <- list(gene2 = c('term1', 'term2'), gene3 = c('term2', 'term3'))
  expect_identical(annots_to_geneGO(d, 'geneID2GO'), e,
                   info = "geneID to GO")
  
  e <- list(term1 = 'gene2', term2 = c('gene2', 'gene3'), term3 = 'gene3')
  expect_identical(annots_to_geneGO(d, 'GO2geneID'), e,
                   info = "GO to geneID")
  
  expect_error(annots_to_geneGO(d, 'bla'),
               info = 'Bad direction')
})

test_that('test_go', {
  # Helper function
  class_and_dimensions <- function(...){
    res <- test_go(...)
    n <- length(res$topgo_res@score)
    res <- topGO::GenTable(res$topgo_data, p.value = res$topgo_res, topNodes = n)
    
    return(list(class = 'data.frame', dim = dim(res)))
  }
  
  # Basic usage
  a <- tibble::tibble(gene_id = paste('gene', 1:5, sep = ''),
                      terms = c('GO:0005575,GO:0044464,GO:0071944',
                                'GO:0044464',
                                'GO:0071944',
                                'GO:0005623,GO:0005886,GO:0071944',
                                'GO:0005886,GO:0016020,GO:0044464'))
  g <- c('gene1', 'gene2', 'gene5')
  e <- list(class = 'data.frame', dim = as.integer(c(6,6)))
  expect_identical(class_and_dimensions(genes = g, annots = a, ontology = 'CC', node_size = 2),
                   expected = e,
                   info = "Basic usage")
  
  # Pass list of annotations
  a.list <- annots_to_geneGO(a, 'geneID2GO')
  # Need to use expect equal because topGO creates some sort of environment
  # for the tests which are never identical.
  expect_equal(test_go(genes = g, annots = a, ontology = 'CC', node_size = 2),
               test_go(genes = g, annots = a.list, ontology = 'CC', node_size = 2),
               info = "Basic usage, pass list of annotations")
  
  # Pass scores
  g.scores <- c(-1, -1, 0, 0, -1)
  names(g.scores) <- a$gene_id
  expect_equal(test_go(genes = g, annots = a, ontology = 'CC', node_size = 2),
               test_go(genes = g.scores, annots = a, ontology = 'CC', node_size = 2, score_threshold = 0),
               info = "Basic usage, pass scores")
  
  # Badd annot object
  expect_error(test_go(genes = g, annots = c('a', 'b', 'c'), ontology = 'CC', node_size = 2),
               info = "Bad annots")
})

test_that("GSEA", {
  class_and_dim <- function(...){
    res <- gsea(...)
    return(list(class = class(res), dim = dim(res)))
  }
  d <- tibble::tibble(gene_id = paste('gene', 1:10, sep = ''),
                      terms = c('term1,term2,term3',
                                NA,
                                'term2,term3,term4',
                                'term3',
                                'term4,term5',
                                'term6',
                                'term6',
                                'term6,term2',
                                'term6.term7',
                                'term6,term2'),
                      score = 1:10)
  e <- list(class = c('tbl_df', 'tbl', 'data.frame'),
            dim = as.integer(c(3, 4)))
  expect_identical(class_and_dim(dat = d),
                   expected = e,
                   info = "Basic usage")
  
  # Increase tests
  e <- list(class = c('tbl_df', 'tbl', 'data.frame'),
            dim = as.integer(c(4, 4)))
  expect_identical(class_and_dim(dat = d, min_size = 2),
                   expected = e,
                   info = "min_size")
  
  d.bad <- tibble::tibble(id = paste('gene', 1:10, sep = ''),
                      terms = c('term1,term2,term3',
                                NA,
                                'term2,term3,term4',
                                'term3',
                                'term4,term5',
                                'term6',
                                'term6',
                                'term6,term2',
                                'term6.term7',
                                'term6,term2'),
                      score = 1:10)
  expect_error(gsea(d.bad),
               info = "Bad column name")
})

test_that('Sign test', {
  # Basic usage. Compare everything except p-values
  d <- tibble::tibble(gene_id = paste('gene', 1:10, sep = ''),
                      terms = c('term1,term2,term3',
                                NA,
                                'term2,term3,term4',
                                'term3',
                                'term4,term5',
                                'term6',
                                'term6',
                                'term6,term2',
                                'term6,term7',
                                'term6,term2'),
                      score = -5:4)
  e <- tibble::tibble(term = c('term2', 'term3', 'term4', 'term6'),
                      n_successes = as.integer(c(2, 0, 0, 4)),
                      expected = c(4, 3, 2, 4) * (4 / 9),
                      n_trials = as.integer(c(4, 3, 2, 4)),
                      p_success = rep(4/9, 4))
  
  expect_identical(sign_test(d, min_size = 2) %>%
                     dplyr::select(-p.value) %>%
                     dplyr::arrange(term),
                   e %>%
                     dplyr::arrange(term),
                   info = 'basic usage')
  
  expect_error(sign_test(as.matrix(d)),
               info = "Wrong input type")
  
  expect_error(sign_test(d %>% dplyr::select(gene = gene_id, dplyr::everything())),
               info = 'Bad column names')
  
  expect_error(sign_test(rbind(d,d)),
               info = 'Non-unique gene_ids')
  
})