Issue when mergerPairs

[Channen1221]

Hello! I am having a bit of trouble and my story is long and irritating...essentially i have 498 samples im trying to run. Im using my pathetic little laptop so running the errors and dereplicating took days and my laptop crashed a few times. So i stupidly went on chatgpt for help. The code i gave me joined the dereplication and dada inference steps. And after each sample, itsaved all my files in a folder (the folder containing the filtered files). I finally managed to get all the data. However, when trying to merge is get an error.

I have all the data saved - the errFs, errRs, dadaFs, dadaRs, filtered files,
(My dadaR and dadaF files are in separate folders)

My code for derep and dada:

for (i in remaining_idx) {

cat("Processing sample:", sample.names[i], "\n")

Dereplicate

derepF <- derepFastq(filtFs[i])
derepR <- derepFastq(filtRs[i])

names(derepF) <- sample.names[i]
names(derepR) <- sample.names[i]

Run dada

dadaF <- dada(derepF, err=errF, multithread=TRUE)
dadaR <- dada(derepR, err=errR, multithread=TRUE)

SAVE immediately (critical)

saveRDS(dadaF, paste0("E:/FastqFiles/dadaFs_", sample.names[i], ".rds"))
saveRDS(dadaR, paste0("E:/FastqFiles/dadaRs_", sample.names[i], ".rds"))

Clean memory

rm(derepF, derepR, dadaF, dadaR)
gc()

cat("Finished:", sample.names[i], "\n\n")
}

Code for merging:

filt_path <- "E:/FastqFiles/filtered"

filtFs <- sort(list.files(filt_path, pattern="_F_filt.fastq.gz", full.names = TRUE))
filtRs <- sort(list.files(filt_path, pattern="_R_filt.fastq.gz", full.names = TRUE))

sample.names <- sapply(strsplit(basename(filtFs), "_F_filt.fastq.gz"), [, 1)
for (i in seq_along(sample.names)) {

• cat("Merging sample:", sample.names[i], "\n")
  

• # Load ONE sample only
  

• dadaF <- readRDS(filesF[i])
  

• dadaR <- readRDS(filesR[i])
  

• # Merge
  

• merger <- mergePairs(
  

•     dadaF,
  

•     filtFs[i],
  

•     dadaR,
  

•     filtRs[i],
  

•     verbose=TRUE
  

• )
  

• # Save immediately
  

• saveRDS(merger, paste0("E:/FastqFiles/merger_", sample.names[i], ".rds"))
  

• # Clean memory
  

• rm(dadaF, dadaR, merger)
  

• gc()
  

• cat("Finished:", sample.names[i], "\n\n")
  

• }

Error in mergePairs(dadaFs, dadaRs, filtFs, filtRs, verbose = TRUE) :
dadaF and dadaR must be provided as dada-class objects or lists of dada-class objects.

Please help!

[benjjneb]

I'd suggest starting to debug by "unrolling" your merging loop, and just walking through the first iteration of the loop by hand.

That is, set i <- 1, then execute each command inside the merging loop. Is that error thrown at mergePairs when i==1? If so, you can than inspect the dadaF and dadaR objects that the error message is complaining about. What are they? That should give some insight into what is going wrong, maybe the readRDS step?

[Channen1221]

I have tried again and now seems to be working. Below is an example of the merges - is this acceptable or quite low?

49 paired-reads (in 7 unique pairings) successfully merged out of 26172 (in 3523 pairings) input.
7 paired-reads (in 2 unique pairings) successfully merged out of 34317 (in 2616 pairings) input.
8 paired-reads (in 2 unique pairings) successfully merged out of 22285 (in 4676 pairings) input.
4 paired-reads (in 1 unique pairings) successfully merged out of 35855 (in 4208 pairings) input.
10 paired-reads (in 1 unique pairings) successfully merged out of 30453 (in 5460 pairings) input.
12 paired-reads (in 5 unique pairings) successfully merged out of 28930 (in 4554 pairings) input.
9 paired-reads (in 3 unique pairings) successfully merged out of 33770 (in 6352 pairings) input.
43 paired-reads (in 7 unique pairings) successfully merged out of 39643 (in 5347 pairings) input.
2 paired-reads (in 1 unique pairings) successfully merged out of 46246 (in 3067 pairings) input.
20 paired-reads (in 5 unique pairings) successfully merged out of 45928 (in 5781 pairings) input.
20 paired-reads (in 5 unique pairings) successfully merged out of 44998 (in 7291 pairings) input.
27 paired-reads (in 4 unique pairings) successfully merged out of 46221 (in 6850 pairings) input.
10 paired-reads (in 4 unique pairings) successfully merged out of 32585 (in 5276 pairings) input.
14 paired-reads (in 4 unique pairings) successfully merged out of 45324 (in 8218 pairings) input.
32 paired-reads (in 6 unique pairings) successfully merged out of 32943 (in 5359 pairings) input.
10 paired-reads (in 4 unique pairings) successfully merged out of 39622 (in 5830 pairings) input.
48 paired-reads (in 13 unique pairings) successfully merged out of 43958 (in 6127 pairings) input.
55 paired-reads (in 7 unique pairings) successfully merged out of 44616 (in 6124 pairings) input.
87 paired-reads (in 6 unique pairings) successfully merged out of 54993 (in 7140 pairings) input.
9 paired-reads (in 3 unique pairings) successfully merged out of 48322 (in 9131 pairings) input.
10 paired-reads (in 3 unique pairings) successfully merged out of 37653 (in 4651 pairings) input.
0 paired-reads (in 0 unique pairings) successfully merged out of 31549 (in 3258 pairings) input.
7 paired-reads (in 1 unique pairings) successfully merged out of 32375 (in 5818 pairings) input.
0 paired-reads (in 0 unique pairings) successfully merged out of 37600 (in 5323 pairings) input.

[benjjneb]

That is way too low a merging rate. My first guess is that you have truncated your reads too much so that they no longer overlap, and thus are failing to merge.

How long is the amplicon you are sequencing? What truncation lengths did you use?