Comparing the overall Kantorovich-Rubinstein distance between truncated and full-length sequencing data.

Quality controlled 454 sequencing data longer than 120 bp was truncated and processed using the pplacer pipeline (Fig. 2). Merged placements from the truncated data for all 60 samples were compared to the same full-length (454 and Illumina) data by calculating the tree-length normalized KR distances between all pair-wise combinations of the three data sets.