GRIMS preferred name
From ICISWiki
GRIMS main > GRIMS Issues and Solutions
Contents |
Setting of IRGC accessions' preferred name
Scenario
Solution
- The general rule is that all germplasm created with a maintenance method should inherit the preferred name from the preferred name of their GPID2 – NSTAT, NVAL, NDATE, NLOCN.
- There should not be two NTYPE 22 for one GID unless they are really alternative accession numbers used by a single genebank. i.e. WARDA was the donor, so WAB 1287 should have NTYPE 22. TOG 5674 should have NTYPE 10 (which means some other genebank accession ID that isn’t the donor’s or the holding genebank’s)
- IRGC accession should not have NTYPE=2 (Cross name)
Communication
From: Sackville Hamilton, Ruaraidh (IRRI) Sent: Thursday, 2008 February 28 10:06 PM To: Prantilla, Roniela (IRRI) Cc: Alcantara, Adelaida (IRRI); Capilit, Grace Lee (IRRI); McLaren, Graham (IRRI) Subject: RE: Setting of Preferred name Ella, Ooooh … too many errors in data as well as in software! I’ll answer the case of TOG 5674, copying to Grace to help correct the PBGB side. Also copying to Graham to demonstrate again the magnitude of the problems we face. I hope Tom Hazekamp will come and help us out. To answer your specific question, the general rule is that all germplasm created with a maintenance method should inherit the preferred name from the preferred name of their GPID2 – NSTAT, NVAL, NDATE, NLOCN. I think this rule is already in Ching’s validation tool. So, if TOG5674 is the preferred name of the GPID2, it should be the preferred name of the IRGC Accno, even if IITA is not the immediate source. BUT: For reasons explained below, in this case I think TOG5674 should not be the preferred name * There should not be two NTYPE 22 for one GID unless they are really alternative accession numbers used by a single genebank. WARDA was the donor, so WAB 1287 should have NTYPE 22. TOG 5674 should have NTYPE 10 (which means some other genebank accession ID that isn’t the donor’s or the holding genebank’s) * The NVAL should be TOG 5674 not TOG5674. We have no reason to use the non-standardised form TOG5674. The original IITA standard is in fact TOg 5674 (with space and with lower-case g). Since their standard includes space, we should definitely not delete it. * The NVAL of the WARDA sample should be the standard WAB 1287, not the non-standard WAB 01287. Different datasets from WARDA variously document it as WAB0001287, WAB 01287, WAB-1287 – as they are not consistent, they clearly don’t think it necessary to write WAB 01287, so we have no reason not to standardise. It is strange to use another genebank’s accession ID as a preferred name. However, if that’s what PBGB wants, let them do it. Since the new ICIS will handle the case of GIDs without a preferred name, you will not need to force donor’s codes to become preferred names. I don’t know how PBGB got their sample of TOG 5674 (GID 537163) (did they get it from us or directly from IITA or even from WARDA?). However, we did not get ours from them, so the GPID2 of ours should not equal their GID, and our accno should not inherit its preferred name from their sample. Here is the history as far as I know it: TOg 5674 was collected by IITA in 19771001 with CVNAM=”EX WURNO”, COLLNO=”OK-90”, from 13°50'N, 05°50'E. It was donated by IITA to WARDA on a date I don’t know and registered under WAB 1287. In the WARDA database, the cvnam is written as WURNO (3), suggesting they think that EX WURNO means selection of WURNO, and assigned different names to indicate different selections. It was sent by WARDA to GRC in 19960226 without info on variety name and registered as IRGC 96790. In the attached excel file, I have summarized the current (wrong) data in IRIS and also shown what should be in IRIS (additions and corrections highlighted yellow). Several errors for every GID and every NID. Ruaraidh
From: Prantilla, Roniela (IRRI) Sent: Thursday, 2008 February 28 13:48 To: Sackville Hamilton, Ruaraidh (IRRI) Cc: Alcantara, Adelaida (IRRI) Subject: Setting of Preferred name Sir, I am beginning to be confused setting the preferred name of IRGC #s. For example, IRGC 96790 have only 2 names and both are of NTYPE=22 (Donor accession #), DONOR_CODE DONOR_SEQ TOG5674 2 WAB 01287 1 The algorithm I developed sets the WAB 01287 as the preferred name since the system recognizes it as the most recent name prior to donating of seeds to IRGC (SEQ=1). And also WARDA is the immediate sender of the seeds, and not IITA. Is this correct? The problem is if we set the WAB as the preferred name, the breeders have already used the TOG in making their crosses. Please see attached communication from Grace. TOG 5674 is the preferred and only name of GPID2. Should the algo then copy the preferred name of GPID2? What is the most logical approach? Many thanks, Ella -------------------------------------------------------------------------------- From: Prantilla, Roniela (IRRI) Sent: Thursday, 2008 February 28 11:09 AM To: Sackville Hamilton, Ruaraidh (IRRI) Cc: Alcantara, Adelaida (IRRI) Subject: Pedigree name in IRIS Sir, Tita Adel created a table called accno_pedigree. Most of them are previously classified as VARNAME in IRGCIS. When we move them to IRIS, what should be their name type? Are IRGC accessions allowed to have NTYPE=2 (Cross name)? Many thanks, Ella
--rherrera 08:21, 29 February 2008 (PHT)
