static void checkXRefRefSeq(struct metaData* md)
/* check sanity of collected additional metadata for RefSeq */
{
if (!md->inRefSeqStatus)
gbError("%s: not in refSeqStatus table, and is RefSeq acc", md->acc);
if (!md->inRefLink)
gbError("%s: not in refLink table, and is RefSeq acc ", md->acc);
if (gbIsProteinCodingRefSeq(md->acc))
{
if (!md->hasProt)
gbError("%s: no peptide for RefSeq", md->acc);
else
{
if (!md->protInSeq)
gbError("%s: RefSeq peptide %s not in gbSeq table", md->acc, md->rlProtAcc);
else if (!md->protInExtFile)
gbError("%s: RefSeq peptide %s not in gbExtFile table", md->acc, md->rlProtAcc);
}
}
}
static void loadRefLinkRow(struct metaDataTbls* metaDataTbls,
struct sqlConnection* conn, char** row)
/* load a row of the refLink table */
{
/* columns: mrnaAcc,name,product,protAcc,geneName,prodName,locusLinkId,omimId */
struct metaData* md;
int iRow = 0;
char *acc = row[iRow++];
char *product;
if (!(startsWith("NM_", acc) || startsWith("NR_", acc)))
{
gbError("%s: non-NM_/NR_ mrnaAcc in refLink", acc);
return;
}
md = metaDataTblsGet(metaDataTbls, acc);
if (md->inRefLink)
gbError("%s: occurs multiple times in the refLink table", md->acc);
md->inRefLink = TRUE;
safef(md->rlName, sizeof(md->rlName), "%s", row[iRow++]);
product = row[iRow++];
safef(md->rlProtAcc, sizeof(md->rlProtAcc), row[iRow++]);
/* check if ids are valid (zero is allowed, so just parse) */
strToUnsigned(row[iRow++], md->acc, "refLink.geneName", NULL);
strToUnsigned(row[iRow++], md->acc, "refLink.prodName", NULL);
strToUnsigned(row[iRow++], md->acc, "refLink.locusLinkId", NULL);
strToUnsigned(row[iRow++], md->acc, "refLink.omimId", NULL);
if (gbIsProteinCodingRefSeq(md->acc))
{
if (strlen(md->rlProtAcc) == 0)
gbError("%s: empty protein acc in refLink", acc);
else
{
metaDataTblsAddProtAcc(metaDataTbls, md);
md->hasProt = TRUE;
}
}
}
static void updateMetaData(struct sqlConnection *conn, struct gbStatus* status,
struct gbStatusTbl* statusTbl, HGID faFileId,
HGID pepFaId)
/* update the database tables for the current entry based on the stateChg
* flags */
{
assert(status->stateChg & (GB_NEW|GB_META_CHG|GB_REBUILD_DERIVED));
/* check for MGC, ORFeome */
if (status->orgCat == GB_NATIVE)
{
if (haveMgc)
status->isMgcFull = isMgcFullLength();
if (haveOrfeome)
status->isOrfeome = isOrfeome();
}
/* clear description if we are not keeping it */
if (!keepDesc(status))
raFieldClear("def");
/* most database changes are only done for GB_EXT_CHG */
if (status->stateChg & (GB_NEW|GB_META_CHG))
{
seqUpdate(status, faFileId); /* must be first to get status->gbSeqId */
gbCdnaInfoUpdate(status, conn);
imageCloneUpdate(status, conn);
}
if (raMiscDiffs != NULL)
gbMiscDiffUpdate(status, conn);
if (raWarn != NULL)
gbWarnUpdate(status, conn);
if ((gSrcDb == GB_REFSEQ) && (status->stateChg & (GB_NEW|GB_META_CHG)))
{
refSeqStatusUpdate(status);
refSeqSummaryUpdate(conn, status);
refLinkUpdate(conn, status);
refSeqPepUpdate(conn, pepFaId);
}
/* update in-memory gbStatus entry */
status->modDate = raModDate;
/* save CDS for use by the alignments */
if (!genbankCdsParse(raCds, &status->cds))
{
/* not valid CDS, only warn if RefSeq, where we expect to be better */
if ((gSrcDb == GB_REFSEQ) && gbIsProteinCodingRefSeq(status->acc))
prWarn("%s: malformed RefSeq CDS: %s", status->acc, raCds);
}
/* geneName for refFlat, if not available, try locus_tag */
char *geneName = raFieldCurVal("gen");
if (geneName == NULL)
geneName = raFieldCurVal("lot");
if (geneName != NULL)
status->geneName = lmCloneString(statusTbl->accHash->lm, geneName);
/* mark as done so dups in other updates don't get added to the database */
status->metaDone = TRUE;
}
