The quality of individual bases in sequencing data is typically assessed through a logarithmic quality scores. These scores constitute a signifi- cant fraction of the total sequencing data storage. As the total volume of sequencing data generated rapidly increases, it becomes important to assess whether the resolution of quality scores can be reduced to alleviate storage requirements.
This white paper examines a method to reduce the resolution of quality scores, enabling a more compact storage of raw sequence reads. Employing a quality scoring scheme with only eight levels of quality or less, the method was tested and found to be virtually loss- less. The analysis results showed no significant differences in variant calling from those obtained with a full quality scale.
