The assembly of human transcripts enabled a greater number of UNIGENE entries with a protein homology. The UNIGENE and HINT consensus sequences were aligned to the SWISSPROT, TrEMBL, and PIR protein databases by use of the program BLASTX. Protein alignments with the highest scores were selected from each UNIGENE and HINT transcript consensus and broken into five groups according to their BLAST scores, ranging from the highest >500 to 100–200 (x-axis). (y-axis) Numbers of transcript consensus. (Open bars) UNIGENE transcripts; (closed bars) HINT consensus.

The data flow in mapping human transcript consensus sequences. UNIGENE clusters were assembled into transcript consensus sequences. The RH mapping information for the individual ESTs and other cDNAs available from Genemap'99 was incorporated into the transcript consensus sequences, along with the cytogenetic and/or chromosomal assignments available in the original UNIGENE data set. Mapping information for each genomic clone was compiled on the basis of the integration of four largely independent maps: Genemap'99, e-PCR and fingerprint maps, and GoldenPath sequence contigs. The resulting transcript consensus was BLASTed against the mapped genomic clones. The final transcript map was substantiated by the integration of multiple maps available for both the transcripts and the genomic clones. (Open circles) Genemap'99 EST RH assignments; (open inverted triangles) UNIGENE cytogenetic or chromosomal assignments; (closed circles) Genemap'99 RH assignments for genomic clones; (closed boxes) additional mapped STSs; (closed inverted triangles) fingerprint-mapping assignments; (bent arrows) GoldenPath positional assignments.

Estimated transcript densities on individual human chromosomes. (y-axis) Putative average number of consensus transcripts per 1 Mb; (x-axis) 24 human chromosomes. The broken line indicates the average genomewide transcript density (∼22 genes per 1 Mb).

The extension of human transcript consensus sequences after the assembly of UNIGENE clusters. The lengths of transcript consensus were compared for each UNIGENE cluster before and after the sequence assembly (log–log scale). The (x-axis) Sequence length for the original UNIGENE transcripts; (y-axis) sequence length for the assembled HINT consensus. Almost none of the anonymous UNIGENE EST entries is longer than 1000 bp (above the limit at regular single sequencing reads), while most of the known genes in the original UNIGENE consensus are longer than 1000 bp.

The database schema for the integration and mining of functional and mapping information of human transcript consensus. The UNIGENE_TBL and Locuslink_TBL were derived from the original UNIGENE and Locuslink data sets, respectively. The Transcript_TBL was created by the integration of UNIGENE and Genemap'99 databases. The cDNA library information in the Library_TBL was standardized further into appropriate categories: Development (e.g., embryonic stages), Cell, Tissue, Organ, Pathology (e.g., tumor), and Treatment. The Assembly_TBL defined the start and end positions and the orientation for each transcript in its contig, and its Smith–Waterman score compared with the consensus. All consensus sequences were stored in the HINT_TBL. The potential splicing variants were represented in the Variant _TBL by the site and length of the insertion or deletion for a given variant. The ProteinHit_TBL and ContigHit_TBL were generated by BLASTing the transcript consensus with the protein (SWISSPROT, PIR, and TrEMBL), and genomic (Ensembl) sequence databases, respectively. A majority of the functional annotations in the Annotation_TBL were derived from the SWISSPROT data set, supplemented with the key words from the PIR and TrEMBL protein entries. The Contig_TBL, based on the Ensembl database, was used to order and position individual sequencing contigs (see Methods). Mapping information for individual clones was integrated by use of four different maps: the assembled genomic contigs (UCSC_TBL); the fingerprint map (FPC_TBL); and the radiation hybrid maps (RH_TBL and e-PCR_TBL). (*) Primary or joint keys for each table; (arrows) one to many relationships; (arrows with closedcircles) one to one relationships.

Estimated mapping error rates by the integration of multiple mapping data sets. A total of 14,469 transcript consensus sequences all containing the following mapping information were tested: the Genemap'99 for the ESTs contained in each of the clusters; the original UNIGENE cytogenetic or chromosomal assignments; the Genemap'99, fingerprint, and GoldenPath contigs (July and October versions), and the e-PCR maps for the genomic clones contained within the transcript consensus. Mapping errors were scored by an inconsistent chromosomal assignment of one map compared with five other maps. Nonspecific BLAST alignments were scored when the transcripts and the genomic clones putatively harboring the transcripts could be consistently mapped to different chromosomes. Errors occurring during the original UNIGENE clustering were detected when two distinct and consistent RH mapping assignments were found within the same UNIGENE clusters. Unconfirmed mapping errors were scored when more than one inconsistent mapping assignment was found for a given transcript.