Metaprocessing in Miqa refers to non-analysis based evaluation of informatics files. For example, all BAM outputs from your pipeline runs are optionally auto-processed to generate statistics including mappability, read length, and error rates. These "metadata" can then be used for evaluation just like any other output information or results -- they can be monitored as metrics, used for benchmarking and concordance, used for rule-based comparison in automated tests, and explored in interactive reports.

Formal Testing

In this example, a simple Folder Stats metaprocessor is used to evaluate consistency on entire output directories of fastqs using a single metric. If exact-match results were not expected or required, these could instead be evaluated by comparing their size, record count, or fastq-stats.

Automation Testing

Similar concordance or benchmark checks can be applied to automation tests, as shown below.

In this example, VCF file outputs for several samples are compared across different pipeline versions. Clicking the summary stat in the table expands a more comprehensive view of all stats gathered by that metaprocessor.