These were rigorously filtered (see supporting infromation) and 353,055 remaining reads (spanning 100,491,819 bases) were assembled using the combined assembler strategy [ 23], employing Roche 454 gsAssembler (also known as Newbler; version 2.6) and MIRA (version 3.21) [ 29] (Additional file 2).
A total amount of 16, 000, 000 short-paired readings (50 60 bases) were assembled.
Based on data corresponding to one single run on the FLX Gene Sequencer from 454 Life Science, almost 100 million bases were assembled into ~53,000 pieces of putative transcripts and the majority of these have been annotated and functionally classified.
Based on data corresponding to a single run on the GS FLX sequencer, almost 42 million bases were assembled into ~43,000 pieces of putative transcripts and the majority of these have been annotated and functionally classified.
After filtering for vector and E. coli sequences, 101,705 reads with a total of 30,549,147 bases were assembled into 803 contigs, 149 of which were > 500 bp and therefore defined as large contigs.
From these data, 847,452 reads (76.6%) with identity tag and length higher than 120 bases were assembled into 42,860 contigs (N50 = 331 bp, consistent with the fraction of the AFLP fragments sequenced: between 200 and 400 bp), leaving 95,812 singletons.
For example, the longest sequences in the P. fastigiatum libraries (10,134 bases, 10,127 bases and 10,229 bases) were assembled using coverage cutoffs three to five and k-mer sizes 25 to 29 while the shortest sequences (< 3,205) were assembled using coverage cutoffs two to seven and k-mer sizes 57 and 63 (see Additional file 2: Table S2 for details).
