comparison test-data/gentest.R @ 8:2cfceb9f9b8e draft

planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/dada2 commit 6248c33a4eacc8785cac321cd4603acf06fae0fd

7:96336c6a2bb7

filt_rev <- c("filterAndTrim_F3D0_R2.fq.gz", "filterAndTrim_F3D141_R2.fq.gz")

print("filterAndTrim")

for (i in seq_len(fwd)) {
  ftout <- dada2::filterAndTrim(fwd[i], filt_fwd[i], rev[i], filt_rev[i])
  b <- paste(strsplit(fwd[i], ".", fixed = TRUE)[[1]][1], "tab", sep = ".")
  write.table(ftout, b, quote = FALSE, sep = "\t", col.names = NA)
}

# In the test only the 1st data set is used
t <- data.frame()
t <- rbind(t, ftout[1, ])

# dada
print("dada")
dada_fwd <- dada2::dada(filt_fwd, err_fwd)
dada_rev <- dada2::dada(filt_rev, err_rev)
for (id in sample_names) {
  saveRDS(dada_fwd[[id]], file = paste("dada_", id, "_R1.Rdata", sep = ""))
  saveRDS(dada_rev[[id]], file = paste("dada_", id, "_R2.Rdata", sep = ""))
}

# merge pairs
print("mergePairs")
merged <- dada2::mergePairs(dada_fwd, filt_fwd, dada_rev, filt_rev)
for (id in sample_names) {
  saveRDS(merged[[id]], file = paste("mergePairs_", id, ".Rdata", sep = ""))
}


# make sequence table
print("makeSequenceTable")

reads_per_seqlen <- tapply(colSums(seqtab), factor(nchar(getSequences(seqtab))), sum)
df <- data.frame(length = as.numeric(names(reads_per_seqlen)), count = reads_per_seqlen)
pdf("makeSequenceTable.pdf")
ggplot(data = df, aes(x = length, y = count)) +
  geom_col() +
  theme_bw()
bequiet <- dev.off()

# remove bimera
print("removeBimera")
seqtab_nochim <- dada2::removeBimeraDenovo(seqtab)

dada2::filterAndTrim(fwd, c("filterAndTrim_single_filters_F3D0_R1.fq.gz", "filterAndTrim_single_filters_F3D141_R1.fq.gz"), maxLen = 255, minLen = 60, maxN = 100, minQ = 13, maxEE = 1)


merged_nondef <- dada2::mergePairs(dada_fwd, filt_fwd, dada_rev, filt_rev, minOverlap = 8, maxMismatch = 1, justConcatenate = TRUE, trimOverhang = TRUE)
for (id in sample_names) {
  saveRDS(merged_nondef[[id]], file = paste("mergePairs_", id, "_nondefault.Rdata", sep = ""))
}
rb_dada_fwd <- dada2::removeBimeraDenovo(dada_fwd[["F3D0_S188_L001"]])
write.table(rb_dada_fwd, file = "removeBimeraDenovo_F3D0_dada_uniques.tab", quote = FALSE, sep = "\t", row.names = TRUE, col.names = FALSE)

rb_merged <- dada2::removeBimeraDenovo(merged, method = "pooled")
saveRDS(rb_merged, file = "removeBimeraDenovo_F3D0_mergepairs.Rdata")

# SeqCounts
get_n <- function(x) {
  sum(dada2::getUniques(x))
}

print("seqCounts ft")
samples <- list()
samples[["F3D0_S188_L001_R1_001.tab"]] <- read.table("F3D0_S188_L001_R1_001.tab", header = TRUE, sep = "\t", row.names = 1)

8:2cfceb9f9b8e

filt_rev <- c("filterAndTrim_F3D0_R2.fq.gz", "filterAndTrim_F3D141_R2.fq.gz")

print("filterAndTrim")

for (i in seq_len(fwd)) {
    ftout <- dada2::filterAndTrim(fwd[i], filt_fwd[i], rev[i], filt_rev[i])
    b <- paste(strsplit(fwd[i], ".", fixed = TRUE)[[1]][1], "tab", sep = ".")
    write.table(ftout, b, quote = FALSE, sep = "\t", col.names = NA)
}

# In the test only the 1st data set is used
t <- data.frame()
t <- rbind(t, ftout[1, ])

# dada
print("dada")
dada_fwd <- dada2::dada(filt_fwd, err_fwd)
dada_rev <- dada2::dada(filt_rev, err_rev)
for (id in sample_names) {
    saveRDS(dada_fwd[[id]], file = paste("dada_", id, "_R1.Rdata", sep = ""))
    saveRDS(dada_rev[[id]], file = paste("dada_", id, "_R2.Rdata", sep = ""))
}

# merge pairs
print("mergePairs")
merged <- dada2::mergePairs(dada_fwd, filt_fwd, dada_rev, filt_rev)
for (id in sample_names) {
    saveRDS(merged[[id]], file = paste("mergePairs_", id, ".Rdata", sep = ""))
}


# make sequence table
print("makeSequenceTable")

reads_per_seqlen <- tapply(colSums(seqtab), factor(nchar(getSequences(seqtab))), sum)
df <- data.frame(length = as.numeric(names(reads_per_seqlen)), count = reads_per_seqlen)
pdf("makeSequenceTable.pdf")
ggplot(data = df, aes(x = length, y = count)) +
    geom_col() +
    theme_bw()
bequiet <- dev.off()

# remove bimera
print("removeBimera")
seqtab_nochim <- dada2::removeBimeraDenovo(seqtab)

dada2::filterAndTrim(fwd, c("filterAndTrim_single_filters_F3D0_R1.fq.gz", "filterAndTrim_single_filters_F3D141_R1.fq.gz"), maxLen = 255, minLen = 60, maxN = 100, minQ = 13, maxEE = 1)


merged_nondef <- dada2::mergePairs(dada_fwd, filt_fwd, dada_rev, filt_rev, minOverlap = 8, maxMismatch = 1, justConcatenate = TRUE, trimOverhang = TRUE)
for (id in sample_names) {
    saveRDS(merged_nondef[[id]], file = paste("mergePairs_", id, "_nondefault.Rdata", sep = ""))
}
rb_dada_fwd <- dada2::removeBimeraDenovo(dada_fwd[["F3D0_S188_L001"]])
write.table(rb_dada_fwd, file = "removeBimeraDenovo_F3D0_dada_uniques.tab", quote = FALSE, sep = "\t", row.names = TRUE, col.names = FALSE)

rb_merged <- dada2::removeBimeraDenovo(merged, method = "pooled")
saveRDS(rb_merged, file = "removeBimeraDenovo_F3D0_mergepairs.Rdata")

# SeqCounts
get_n <- function(x) {
    sum(dada2::getUniques(x))
}

print("seqCounts ft")
samples <- list()
samples[["F3D0_S188_L001_R1_001.tab"]] <- read.table("F3D0_S188_L001_R1_001.tab", header = TRUE, sep = "\t", row.names = 1)