Preprocess report overcounting low-complexity reads #592
Description
Running sunbeam v5.2.1 with slurm profile and conda install.
The preprocessing report table results in values in the 'komplexity' column about 30-40% higher than the difference in read counts in the fastq files. This can be attributed to duplicate lines in the log/komplexity/[sampleid].filtered_ids files. This may be that the software is trying but failing to drop just one read of the read-pair? The RP info does not show up in the filtered_ids file.
