Figure 6.

Hundreds of transcripts assembled de novo do not align to the reference genome. (A) Number of unaligned contigs assembled by Trinity at every developmental stage. The portions shaded in gray represent the contigs that have EST support, while the unshaded portions correspond to the contigs that do not match any existing Xenopus EST and are not derived from contaminating sources. (B) Number of unaligned Trinity contigs plotted against sequencing coverage. A linearly increasing trend is observed, indicating that deeper sequencing is required to discover even more transcribed sequences that are missing from the reference genome. (C) Validations of unaligned Trinity contigs with EST support. All the PCR products were sequenced to confirm that they were indeed the relevant contigs. The names in brackets correspond to the matching ESTs or cDNA clones, which were previously sequenced as part of either the NIH Xenopus Initiative (Klein et al. 2002; Gerhard et al. 2004) or the Sanger Xenopus tropicalis EST/cDNA project (Gilchrist et al. 2004; Carruthers and Stemple 2006). (D) Validations of unaligned Trinity contigs that have never been detected prior to this study. All the PCR products were sequenced as confirmation. The name of each contig was arbitrarily given by the program. One contig, comp29928, was most highly transcribed from Stages 13–28, while the expression of the other two contigs, comp697 and comp39806, showed a general increase over development and remained strong even in the feeding tadpole stages.