We identify gaps in RNAseq genelet data and run PCRs to bridge them. Using RNAseq aggregate integrated genelets, we construct a set of merged genelets from genelets that are not known single exon transcripts by forming the unions of genelets with overlapping exon bases. The terminal genelet features consisting of transcript start sites, SL acceptor sites, transcript end sites, and poly-A sites are assigned to a merged genelet with the genelets that compose it. For each chromosome and strand, we select candidate genelet pairs in which neither facing genelet end has an assigned terminal feature, the gap is up to 5000 bases long, and the genelet pair has average
coverage per base of 50 or less where the genelet pair coverage is the value for the lower coverage genelet. After removing candidate pairs that are covered by existing data, we pick PCR primers for the remaining candidate pairs so that the product contains at least one known intron. The PCR products are sequenced from the PCR primers using a capillary DNA sequencing machine. We use phred to call bases and cross_match to align with splicing the resulting reads to the genomic sequence. Genelet gaps are considered to be closed when the alignments cover at least the genelet bases at the gap ends.