Which trimming pipeline is implemented on the calculation engine when a read-based job is submitted?
de novo assembly and reference mapping:
When a de novo assembly or reference mapping job is submitted to the calculation engine, an automated trimming procedure is performed on the calculation engine before the actual job is started. In the first step of the automated trimming pipeline, the read statistics are calculated. Based on these statistics, read set estimates are calculated and used to derive the actual trimming parameters. Based on these parameters, the trimming algorithms (quality trimming, structural trimming and read length trimming,) are executed on the reads. The quality trimming is done with a minimum quality threshold of 10% of the average quality. The structural trimming removes low quality end, with a rolling average quality threshold of 25% of the average quality. Finally, all reads below 10% of the average read length are removed. The resulting trimmed read data set is used as input for the selected job(s). The quality statistics of the trimmed dataset are imported in the BioNumerics database and stored in the “NameExp_trimmedStats” experiment.
assembly-free calling:
The assembly-free calling algorithm on the calculation engine includes a trimming and quality control check. Every indivual k-mer is checked for base call quality and forward, reverse and total coverage.