Below is my response to a thoughtful commentary by Greg Shaner (which follows my comments). This is a critical topic and I'd appreciate any input anyone can give.
Hi Greg (and everybody else): Thanks for your comments. Your logic is sound, but I'm going to argue that it is the logic of a "study" based approach versus what I'll call a technology employment approach. I see markers as having three primary utilities. First, 1) they can 'tag' useful loci/alleles. This requires some sort of controlled mapping population (RI, BC, etc.) that is both genotyped and phenotyped. Second, 2) once associated with a useful gene, markers can be used to identify segregants carrying the targeted gene. Thirdly, 3) they can be used to accelerate the rate at which unwanted genome segments are discarded in introgressive breeding systems (like backcrossing, either formal A/B//A, or adapted/unadapted(but with good gene)//adapted2.) I think the approach you outline is very valuable, but I think it only addresses the first application of markers. Tanksley's presentation at the ASA meetings highlighted an approach that simultaneously does all three activities and has generated commercial NILs with a very small number of backcrosses. We have more linkage groups , more difficulty creating seed for large populations, and less robust marker technologies, so I don't quite expect that sort of success. Kim and Fred both saw that emphasis should be placed on number one and number 2. In the case of #2 (id favorable segregants) they would do that within normal breeding populations, and for #3 they would use adapted/unadapted//adapted crosses. I'm really concerned that the approach you outline is much to limited in its power to exploit positive alleles from other sources. We already have a half dozen alternative sources of resistance (european, chinese, frontana, freedom, etc.). I'm confident (without any evidence at all!) that we will find more as we engage in a systematic search.
When I asked the question, 'is now the time to make the jump into molecular marker facilitated breeding', I wasn't asking if we should expand our efforts at gene mapping (#1 above) only. I meant is it time to fully employ this technology, which I foresee involving all three activities. We will simultaneously tag useful genes from new sources of resistance; employ markers for known genes to restrict our breeding materials to segregants with positive alleles, and accelerate the reassembly of adapted or recurrent parent genome segments in introgressive approaches. This is going to be as integral a component of breeding as yield testing and disease screening are today. The question is, are we at that point now?
At 12:41 PM 3/14/97 EST, you wrote:
Received: by TAOMLR3@MSU ; Fri, 14 Mar 97 12:39:14 Received: from MSU (NJE origin SMTP@MSU) by MSU.EDU (LMail V1.2a/1.8a) with BSMTP id 0717 for 22857MGR@MSU; Fri, 14 Mar 1997 12:39:12 -0500 Received: from herald.cc.purdue.edu by msu.edu (IBM VM SMTP V2R3) with TCP; Fri, 14 Mar 97 12:39:09 EST Received: from mailcenter.btny.purdue.edu by herald.cc.purdue.edu; Fri, 14 Mar 97 12:38:50 -0500 Message-ID: n1353779367.89684@btny.purdue.edu Date: 14 Mar 1997 12:36:16 -0500 From: "Greg Shaner" shaner@btny.purdue.edu Subject: Proposal To: "Patrick Hart" 22771lph@msu.edu, "Rick Ward" 22857MGR@ibm.cl.msu.edu Cc: "Robert Todd" rtodd@badlands.nodak.edu X-Mailer: Mail*Link SMTP-QM 4.0.0
Rick and Pat, Sorry to hear you had trouble getting back to East Lansing yesterday. Bill and I drove through heavy rain all the way back to Lafayette, but we still made fairly good time. I think the meeting yesterday was productive. Thanks to the two of you and Ian Gray for taking the initiative on this. (Dr. Todd, I am also pleased you were able to attend. Your advice on how to handle these kinds of proposals was very helpful and kept us from getting to fanciful.)
I have had some time to ponder what we decided, and have a couple of thoughts about the genetics and breeding aspects of the program. I am taking today off, and only came in during the noon hour to see if you had sent a draft proposal, but from your note to Jackie, it sounds like you have not had time to do this after your travel adventures. So, I will send you some ideas that you might incorporate into the document. I will probably come up to my office tomorrow or Sunday sometime, and will check my email then to see of a draft proposal is available. Anyway, my thougts about the breeding/genetics program are as follows:
Since right now everyone is working with basically the same resistances (Sumai 3/Ning 7840 or Frontana) it doesn't make scientific or financial sense for every state to get $125,000 for marker work. That looks like a lot of duplication of effort. The first step should be to take a couple of mapping populations, perhaps Ning 7840 x Clark and Sumai 3 x Stoa and have those phenotypically evaluated at several locations. Even though these evaluations would be in the greenhouse and use the same inoculation technique (single floret inoculation), it is important to see if each lab would find the same degree of resistance in each family. This would provide critically needed data on genotype x environment effects. Assuming the multi-location trials give essentially the same results, then mapping could proceed on these populations with a high degree of confidence in the phenotypic classification. The mapping itself should be done in perhaps only a couple of labs. I am not competent to say whether AFLPs are the way to go, or some other technique should be tried too. Once linked markers are found by these two labs, the markers should be made available to all labs who want them, and used to follow segregation of resistance in crosses between Sumai 3 or Ning 7840 and local, adapted, susceptible cultivars. This would provide the final proof that the markers are indeed tightly linked to the genes for scab resistance, and then breeders could start using them routinely. As other sources of resistance are identified, they should be crossed to Ning 7840 or Sumai 3 to determine, by phenotypic classification, whether they carry unique genes. The markers could also be used to determine if these new sources of resistance carry resistance genes at unique loci. When unique sources of resistance are found, then genetic studies should proceed, to characterize the inheritance of this resistance and look for markers for the genes. Again, this preliminary marker work should be done in only one or two labs, not necessarily the ones that did the work on Ning 7840 and Sumai 3. Maybe whoever finds a new source of resistance should have first crack at the marker work. I don't think the proposal needs to go into all this detail, but it should not show that every state gets the same share of the mapping money. This looks like we are just trying to make everyone happy instead of really setting up a regional cooperative research program. Perhaps the proposal could state the total amount needed for mapping, then say that phenotypic classification would be done on replicated mapping populations at several locations (perhaps all locations) to measure G X E and to make sure each location can obtain results that are consistent with all others (sort of like standardizing labs that do regulatory work), but that two states would get extra money out of the mapping pot to do the actual screening of markers. As for breeding, we can make a case that every state should get an equal share. Scab is sufficiently important that every variety released from now on should have resistance. It is not sufficient to have a few resistant varieties available -- every variety, regardless of its other virtues and defects, should have scab resistance. Therefore, every state that has a breeding program (i.e. all 8 states represented yesterday, and KS and NE if they want in) should get an equal share of the breeding money. Once markers are available, some of this money can be used for routine use of markers to screen segregating populations.
Regards, Greg shaner@btny.purdue.edu
Rick Ward
Associate Professor/Wheat Breeder Michigan State University wardri@pilot.msu.edu (517)-355-2231(land) (517)-285-9725 (cell) (517)-353-3955 (fax) address: MSU/Dept of CSS 382 PSSB East Lansing MI 48824 USA