This file was obtained following this steps:
Download from http://hgdownload.soe.ucsc.edu/goldenPath/hg19/database/ the refGene.sql, knownGene.sql, knownToRefSeq.sql, kgXref.sql tables
Import everything it in mysql
Generate views for annotation:
`chrom`, min(kg.cdsStart) as CDS_minStart,
SELECT kg.max(kg.`cdsEnd`) as CDS_maxEnd, k2a.geneSymbol,
`strand` as GeneStrand, min(kg.txStart) as TSS_minStart,
kg.max(kg.txEnd) as TSS_maxStart,
kg.proteinID as ProteinID, k2a.protAcc as ProteinAcc, k2a.spDisplayID`knownGene` AS kg JOIN kgXref AS k2a
FROM = k2a.kgID COLLATE latin1_bin
ON BINARY kg.name -- latin1_swedish_ci
-- WHERE k2a.spDisplayID IS NOT NULL and (k2a.`geneSymbol` LIKE 'Tcra%' or
`geneSymbol` LIKE 'TCRA%')
k2a.WHERE (k2a.spDisplayID IS NOT NULL or k2a.spDisplayID NOT LIKE '')
`geneSymbol` LIKE 'Tcra%'
and k2a.`chrom`, k2a.geneSymbol
group by kg. ORDER BY kg.chrom ASC , kg.txStart ASC