In the Quality Control Step of a typical NGS post-sequencing pipeline, we used Trim Galore! for phred score trimming of reads. As I understand it, Trim Galore! trims the ends of a read should they fall under the specified phred score threshold but does not remove bases which may fail the threshold should they be within the read (from Trim Galore! –help: The algorithm is the same as the one used by BWA (Subtract INT from all qualities; compute partial sums from all indices to the end of the sequence; cut sequence at the index at which the sum is minimal)). As a result, reads potentially contain bases with phred scores under the threshold (tested and shown in our data).
During methylation calling, we note that some softwares further filter for base phred score before methylation calling (Strand NGS, MethylExtractor), whereas others recommended for deduplication before methylation calling (Picard, samtools) do not. Filtering per base phred scores can potentially heavily influence the outcome methylation call (also tested and shown in our data). Agreeably should a significant quantity of reads contain low phred scores in the middle of the read, it would be reflected in various modules of FastQC – however, our question is, is it necessary to further filter by per base phred scores after an initial read phred filter by Trim Galore! and no major flags by FastQC?
During methylation calling, we note that some softwares further filter for base phred score before methylation calling (Strand NGS, MethylExtractor), whereas others recommended for deduplication before methylation calling (Picard, samtools) do not. Filtering per base phred scores can potentially heavily influence the outcome methylation call (also tested and shown in our data). Agreeably should a significant quantity of reads contain low phred scores in the middle of the read, it would be reflected in various modules of FastQC – however, our question is, is it necessary to further filter by per base phred scores after an initial read phred filter by Trim Galore! and no major flags by FastQC?
Comment