ReadTracking seems broken with single sample runs

[cjfields]

ReadTracking seems to be broken with the test data:

Command exit status:
  1

Command output:
  [1] ‘1.14.1’

Command error:
  Loading required package: Rcpp

  Attaching package: ‘dplyr’

  The following objects are masked from ‘package:stats’:

      filter, lag

  The following objects are masked from ‘package:base’:

      intersect, setdiff, setequal, union

  Error in getUniques(x) :
    Unrecognized format: Requires named integer vector, fastq filename, dada-class, derep-class, sequence matrix, or a data.frame with $sequence and $abundance columns.
  Calls: as.data.frame -> sapply -> lapply -> FUN -> getUniques
  Execution halted

Work dir:
  /home/groups/h3abionet/Microbiome-Tests/results/2020-09-26-issue10/work/a5/b0e86df104462a14845090fdf5b983

Tip: when you have fixed the problem you can continue the execution adding the option `-resume` to the run command line

The error appears due to some changes with the getUniques function and the classes returned from the learnErrors calls; the current implementation expects a list but the returned values are single objects. I suspect there was an API change that may be causing this

Oddly this does work with our recent local runs for 16S so will require a little investigation as to why.

[cjfields]

Looks like this is due to the same issue as #1, our local config was running one test pair unintentionally. I'll go ahead and try debugging this but will close the first ticket as it's not as informative.

[cjfields]

The issue can be traced back to the derepFastq function, which gives a single derep object with one input but a list of derep objects with multiple inputs. The call in the ReadTracking process expects a list.

[cjfields]

This seems largely fixed now and was related to how tables for single samples must be forced in R.