Possible bug in PRODUCT_HEATMAP (Pandas)

[shaze]

Description of the bug

ERROR ~ Error executing process > 'WRIGHTONLABCSU_DRAM:DRAM:PRODUCT_HEATMAP'

Caused by:
  Process `WRIGHTONLABCSU_DRAM:DRAM:PRODUCT_HEATMAP` terminated with an error exit status (1)


Command executed:

  dram_viz --annotations raw-annotations.tsv --groupby-column input_fasta

Command exit status:
  1

Command output:
  (empty)

The error is this

ValueError: failed to validate FigureOptions(...).y_range: expected an element of either Instance(Range), Tuple(Float, Float), Tuple(Datetime, Datetime), Tuple(TimeDelta, TimeDelta), Seq(String), EagerSeries or PandasGroupBy, got [4915, 4916, 4917, 4918, 4921, 4922, 4923, 4924, '4924', 'C178', 'C178B', 'C179', 'C180', 'C181', 'C181B', 'C182', 'C183', 'C184B', 'C185', 'C186', 'C186B', 'C187', 'C188', 'C189']

The 4915 etc are the names of our samples. The problem seems to be that the samples that are numbers confuses the typing of Pandas when it reads in column 2 of the tsv file.

If I replace the 4915 etc with S4915 etc, the code runs.

System information

DRAM : DRAMv2 [756904d]
Nextflow version (eg. 23.04.0)
Hardware: HPC
Executor: SLURM
Container engine: apptainer 1.4.3-1.el9
OS Rocky 9.6

Command used and terminal output

Relevant files

No response

[madeline-scyphers]

Thank you so much. I am pretty sure this is an easy fix where the pandas read_csv code is just inferring type, and I think it needs to explicitly cast it to a string. Will get working on that!