An alternative hypothesis is that duplicated sequences within the rhesus assembly were not assembled in the chromosome scaffolds and were instead placed in the "chromosome unknown" or chrUr bin.
a The genomic regions which were not assembled in 454/Draft assembly are listed above; b the 'x' coverage defines the raw-read coverage averaged over given coordinates; c 'Complete/partial' contig coverage defines whether the region was completely/partially assembled while 'None' defines that this region is missing in the respective assembly.
Roughly 94 96% of the chimeric sequences were either split correctly at the chimeric junction during assembly, or were not assembled at all.
Despite generating an efficient reference assembly, over 66 million channel catfish reads and 90 million blue catfish reads were not assembled with the reference assembly.
Depending on the sequence quality and the criteria specified for filtering the data prior to assembly, the samples were not assembled.
The missing sequences corresponded mainly to transposase regions which were not assembled possibly due to the assembly and clustering procedure which often has difficulties reconstructing repeated variable regions.
Reads that were not assembled into contigs in the reference assembly were entered into a subsequent de novo assembly with a higher stringency minimum match similarity (90%).
Pseudo-singletons refer to the de novo assembled Illumina sequences that were not assembled into contigs during the Gene Index Assembly process.
All the newly generated expressed short reads were first assembled using reference assembly (Table 2), and those that were not assembled, i.e., they did not align in silico to the existing catfish ESTs, were used for the de novo assembly.
In our assembly, 8% of the quality-filtered reads were not assembled [similar to [ 31, 38]], and these singletons were merged with the contig set into a preliminary EST set.
They were not assembled casually.
