 Good afternoon everybody. I am very honored and pleased to introduce our assembly speaker for today. He's Dr. Bert Collier, as you see here. He's our senior scientist in the division. And he joined E.V. in March 2011. But before that, he was also at E.V. like many of us here and then returned back in 2011, or 2011. And joined our irrigated rice breeding. Now he's leading it. And before that, I would like to mention that he was at the University of Melbourne when he did his PhD and his bachelor's degree with honors. And from E.V. in 2005 to 2006, he went back to Australia and joined the primary industries in the department of primary industries in Australia and became the national program leader of the Durham Wheat breeding program. So here at E.V. he's leading our irrigated rice breeding program. And I would like to introduce him as our speaker for today, Dr. Bert Collier. His talk will be on redesigning the E.V. irrigated breeding program. Dr. Collier? Good afternoon to us. So I think many of you are aware that there's been an enormous amount of changes happening in the plant breeding and genetics and biotechnology division. It's probably an understatement because I dare say that there haven't been such dramatic changes in the last 50 years. I will tell you this afternoon about just some of those changes, particularly concerning the irrigated breeding program. So it's too much. One Thursday seminar is not enough to talk about all of them. And I will leave some content for my colleagues to inform you about in the future, but I will specifically talk about the irrigated breeding program. First, we might start with just a little bit of history. I think most of us are aware that E.V. has a very illustrious and rich history. It really started in 1966 when IR8 was developed and released. And I think sometimes sobering to remember that at the time, farmers used traditional varieties and typical yields were several tons per hectare. But after this first semi-dwarf variety, the yield was shot through the roof. We're talking six to eight tons per hectare. And even in some really favorable environments, it would nearly break 10 tons per hectare. So this was truly a revolutionary change in rice production. I had to do some counting to see how many varieties that had actually released. From the irrigated program alone, there were 45 or something like that. It sounds hard to keep track of all of them because sometimes varieties are released in different countries with different names. But it's really a credit to the breeders and the breeding team that really developed all of those varieties. So initially from Dr. Kusch, Dr. Pamina Verk, and the team, the local staff that really spent so many long hard hours in the sun. But it's not the quantity, it's really the quality. And what I mean by that is that some would argue the true measure of how successful the variety is is certainly not, you know, one doesn't measure how good a breed is and how many varieties are released. It's really what is done with them. And I've just selected two varieties which really almost led to the term, a new term to be proposed called mega varieties. And I think Dave McKeel was the first person to describe this term. It means a variety that's grown on millions of hectares. So we're not saying it's just popular, it's extremely popular. And if you look closely here at this table, just for IR 36, that was arguably one of the first mega varieties of rice that in the Philippines alone, it was pushing 4 million hectares. But the real superstar of rice varieties was undoubtedly IR 64. And if you just look at the coverage of this single variety alone, it is quite staggering. Just in Indonesia, 47 million hectares, it's quite remarkable. And I'm not aware of any other crop that even comes close. For our friends from SSC, this is probably the bottom line. There was an independent assessment done by ACR, which is the Australian Centre for International Agricultural Research. And they estimated the impact of irrigated varieties in three Southeast Asian countries. And that was the bottom line there. 1.46 billion dollars in terms of the economic benefits per year. And they broke it down per country. And you can read this report for yourself. So it's a free, it's actually downloadable from the ACR website. So, with that rich history and that quite grand history, in fact, some have described IRRI's irrigated program as the most famous in the world. And when you consider what the outputs have been, that is probably a fair comment. So the question you're probably asking is why have we changed things? I shall explain. But first, I'll just give you some background which I think is critical for the context of what I would say. First point, this is like plant reading 101. First point, just so you're all on the same page, rice breeding takes a long time. It's generally accepted here at IRRI that a typical time to develop a new variety would be about nine to ten years. So it's a lot of work. Actually, it literally is a lot of work. From seeding to harvest, there are just an enormous amount of that operations that are required in the field in front of the computer, storing seeds and what have you. There's an enormous logistical operation. Those in TVGB know this only too well. Another point I'd like to emphasise is that, by the way, this book, it's a free IRRI book which you can Google. And I used to basically give it to the participants of the rice breeding course. That was the first thing I would encourage them to read. It's really the Bible of rice breeding. I still sometimes refer to it from time to time. It's a beautiful textbook. But the thing I'd like to point out, because you may not be able to see it at the back, is its publication date, 1979. And if one was to look at it, you would probably see that little here at IRRI and that really represents what Asia is on about has changed in quite a long time. So that's the background. So why have we changed? Now I'll explain why the reason. The first one, you have to pardon my language if I get a bit too rough, but I suppose the breeders are a bit like this. I call it the F word. That was our main reason. Almost overnight, because of full cost recovery, we had to change our operations. This just gives you a summary of the amount that was spent. Actually in 2010 and 2011, full cost recovery wasn't implemented. So I basically got the samples that were in our records and did a back calculation in our current full cost recovery rates. But you can see here that just, this is just the trade screening for disease and second quality alone. This does not include our field costs and all of our other labor resource costs that are required. And maybe it will just change some life, but our main funding source in the last five years from 2009 to 2014 has been the Japan Rice Breeding Program. And you can see that the cost of screening alone exceed what we're allocated from that program. So reason number one is from the point of view of resources alone, I believe we have to change. And you'll notice also in 2013, there was no screening done because I had a deficit to carry over for that year. I had no money to do anything with full screening anyway. So that's the first reason. I think it's also with acknowledging the flaming support from the Japan government for the Japan Rice Breeding Project. Reason number two, there have been tremendous changes within the division as I mentioned just before. In the last decade and a half, I think we were something like this. There was a very big irrigated program and that included the golden rice component. There was a hybrid rice, which is still pretty much the same now. And I think we were kind of based in these great-based rain-fed focus breeding programs. And really the strasa program has just supported those pipelines. And there was always some japonica work that was done. 2012, 2013, we were transformed into this. I think many of you would be aware of the level of restructuring that's been done from our head, Iroh and Asila. And what we look like now, so I mean, it's similar, but it is certainly different. We've got a focus on irrigated and rain-fed ecosystems and we've really been divided a lot more into regions, as you can see from this diagram. I will also highlight, for those of you who may be not aware of it, that we've also really decentralized a lot of our breeding activities as well. So, for example, the hub, as we call it, for South Asia and irrigated, is located in Hyderabad in India. And the same is for the rain-fed program, which I've in Kamali in Hyderabad, same location. And there is also another hub in Burundi. So that's a huge change in terms of just the way the breeding pipelines as they're called are organized. Many of you are probably aware that, for example, there used to be a drought breeding program. There still is, that is certainly going on, but the way we've been restructured has been focused on generally separating the trade development work or the research work, or even some people might look pre-breeding work from the actual breeding itself. And so that's been done for most of the traits, if not for all of the traits that Irie was working on. So one representation might look something like this, that there's still a lot of input that they had into the breeding pipelines, but they are separated. And the way I like to sort of think about it is, all of them still flow into the relevant breeding pipeline, the variety development pipeline. So, and as you can imagine, some trade teams actually flow into multiple pipelines. So that's how the structure is. But again, I'm not gonna talk so much about PBGB itself, I'm gonna focus on the irrigated activities. Reason number three, undoubtedly concerns with just a change of philosophy. And I simply mean just new ideas about the best approaches, using modern methods, using alternative methods. And as you saw from the first, the introduction slide, I mean, my background is from other crops, and I've seen things done completely differently with cereals like wheat and barley, which are very similar to rice, but of course they're breeding in a very different way. Reason number four is the high demand on iris scientists. And I'll actually mention breeders here. I don't think we discussed this much, but compared to a typical NARS breeder, we have a lot more on our plate. We have seed requests. We have, and just this year alone, I mean, we're only August, right? I've had, we have, our teams have thousands of individual seed requests, and it's been quite demanding to say the least, to fulfill, and we're obliged to do so. And we want to do so, that's what iri's about, but it has been a challenge. We have visitors nearly every week, every day of every week. We have training courses. I've just come back from a mission in Vietnam, and I'm still burning off the excess food that I ate during that time. We have all sorts of research activities, and I think it's fair to say that, I think breeders need to be aware of that. And my point is, I think there's no harm to simplify methods whenever possible. And one thing I always like to give, I'm not sure if he's here, but I'd like to acknowledge Alan and the INJA team for really facilitating those hundreds and thousands of entries that we send to our NARS programs every year. Okay, reason number five, this is possibly the heart of the redesigning activities that I've come to mission. This is a figure from, I think one of the landmark papers, it's an absolute beautiful paper from Xiaobing Peng and others here at Iriff. And it's quite simple. It just basically describes what we refer to as an era experiment. So basically you take the varieties from the old, the oldest varieties that you've got, and you include the current and older ones, and you put all of those and benchmark them with the modern, the latest varieties. And that's what that was done. So it was published in 1999. And I think this one here turned out to, it was an elite line at the time, but it turned out to be PSBRC 52. So that was the data set that they had at the time. And the bottom line was that they estimated that the rate of genetic gain or yield improvement over time was about 1% per year. Okay, that's the first point. And if you look at the, if you read the paper more, that translates to about 75 kilograms per year, per hectare per year, just to give you a sort of a reference point. Based on Erie's projections and a lot of other data, this is believed to be insufficient to meet future demands for rice. So this is something, a very important take home message. We've actually extended this experiment. As you saw, there are only about 12 varieties that met in the previous one. And that was done 15 years ago, or even further than that. And so what we did, this was one of the first things in 2013, we started to basically reassess where we are at this, or in 2013. And so we put all of the varieties that were included in Shelby Paine's paper, including IR8, and we put the latest ones there. And we found out that, I should point out, I should have mentioned in the previous slide that that study was done during the dry season when yields are higher, of course. But we wanted to test both seasons. And this is still ongoing, but we found that perhaps we feel that the estimated rate of genetic gain is actually lower than first predicted by Shelby Paine's paper. And the rate of, you can see there's a lot of noise in this graph, but certainly the rate, we certainly believe for the wet season at least, it is lower than what was first reported by Paine. Okay, so on the basis of that, if you wanted one more reason, reason number six is, as I said, the Japan Britain program money was to end this year, and we had to look for an alternative source. And the Gates Foundation was interested in some discussions with IR8 about our new ideas to change that last point, the rate, improving the rate of genetic gain. And so reason number six is really, we got money to do things differently. We did not get money to do things as they were done previously. And I think that previous slide shows about 50 years of our best efforts, and the rate was quite linear, it certainly was linear. And that's, as I said, we really feel that we need to do more than that. So I'll introduce this new project, which we call Transforming Rice Breeding, or TRB. It's a Gates Foundation project, like many others at the moment. Our total amount is 12.5 million over the five years, and it's specifically focusing on irrigated regions. And it's not specifically our pipeline, but it really supports a lot of transformational activities within the PBGB division itself, and I'll elaborate more during that. But certainly more will be said about in the near future. And just to complete the picture, that the focus of this research project was exclusively South Asia, namely India and Bangladesh, which are priorities for the Gates Foundation. Okay, I'll start by talking about breeding methods, and this is also the most controversial, so I think it's a good place to start. Again, plant breeding 101. This is a simplified version of the pedigree breeding method, which has been used for the last 45, 50 years plus, actually, in the irrigated breeding program. And probably the basis of this is that there is extensive trade selection during the segregated populations that, so the new breeding populations that are developed, they're screened for many traits and combinations of traits at the same time as well. So we've got here, last, and there's several variations possible, but this is just one scheme. So that's why it was costly and even time-consuming and resource-demanding because there was so much trade screening done and individual lines had to be kept intact in terms of their identity from a single F2 plant. What we're trying to do is to use a new approach it's called Rapid Generation Advance, or RGA. It actually isn't that new. It was first reported in the, I think it was the 1940s, can you believe? But it really, it was adopted by some programs, particularly winter cereals, made in the 60s and 70s. So it's not that new, but probably the bottom line is that you grow plants in these little seedling traits at high density with low fertilizer. And in fact, I think Ricardo, I think I'm going to get a prize for the first crop of the year. I will show you that also, RGA trait. I hope you can all see that. So the features are that, yeah, you just basically grow plants at a very high density and you stress them. You actually impose a mild stress. You can see that if this was in the field, this would be tilting quite extensively. But in these small traits where we can find the growth of the plant, it still is not possible because the plant is stressed. And when the plant is stressed, we manipulate the conditions in such a way that it will flower much more early than it would if it was grown in optimal conditions. The other point I'd like to emphasize is that in contrast to the previous method, there is no trait selection during what we call line fixation or line development stage. Okay, so that's quite a big difference. Although, as I said, this method is not actually that new. It's been used in a lot of other breeding programs, but it's probably had a limited use in rice, but some use in rice in fact. This is just before Typhoon Glenda, we had a really nice operation in VGO 9 and it looked something like this. We had a two-tiered shelving system, so space was our main limitation. And you can see probably the thing I just wanted to highlight by showing this short video is we've got a lot of plants in this small area. That's probably the take home message. And that was in March, April, early this year during one of our first round hits, Joseph Boreto, who I'll say more about him in a moment. We had about 40,000 lines in what you just saw, and that greenhouse wasn't even full. The equivalent area in the field would take several hectares, okay? So this is currently where we're at. We're still doing a lot of optimization because rice is actually not as easy to manipulate as, say, wheat or barley. But we've made excellent progress and we've got several experiments planned or even some that are ongoing. And what the idea is, is to make this a service, a full cost recovery service, so that other breeders, other trade teams can actually use it. And I'll target some time in 2015, really, as I said, space is our limitation and so new facilities are being planned. So we've actually taken a few guinea pig customers and we're seeing how that goes because the real challenge is the wet season. And let's hope there are no more typhoons this year but the wet season brings a load of insects particularly which once they're in, greenhouse can really wrecks and damage. And for those customers or for further information, if you have any complaints, these are the people I'd like you to contact, not me. Joseph and Rulix will be leading the RGA team and they've made an excellent start to this very critical stage of our new cross-cutting team. And it's probably worth pointing out that probably only 18 months ago to two years and we had just basically just a concrete area, there was nothing, we had no buckets or we had nothing. So we've come a long way in a short time. Some of you might have seen this sign. What that's about is, well, space really was limited and we just thought, well, would this work in the field? And we thought, well, why don't we just try and just stick some in the field? So we literally did that and at first we were thinking, well, can we just basically grow them in the field? But the problem, as you can imagine, is the tillering and the resources, the nutrients that the rice plant would have. So by keeping them in this tray, we still confine the growth and we still maintain that low level of stress on these rice plants. And so this started in May, this is the first season that we've run it. And if you just walk past it, I think you would have seen it because it's next to the long-term continuous cropping experiment. It really looks like a sort of a seedling nursery that's been left to run for too long, okay? So you can barely notice, but the trays are there. All this material has been planted direct seeded into trays and just literally stuck in the soil. And just yesterday, this photo was taken just yesterday afternoon, they've just started to harvest that material. And because it's direct seeded and there's that moderate stress needed for RGA, I counted the calendar days between seeding to the start of harvest, it was 91 days yesterday. And I think we can tweak this system a lot more. How far we go with it, I don't know, but certainly in the meantime, when we don't have enough space in the greenhouse, especially after the typhoon, I think we'll run with this. We certainly need to tweak the fertilizer and maybe some other management, but it's promising. And again, in half of the area, which is, it's about the total area is something like 400 square meters. We had over 260, that's a predicted amount, 260,000 lines in about 200 square meters. So it's quite a really densely packed area. And I should have also mentioned that since we are neighbors to the long-term continuous copying experiment, we're also following their seasons. So they have the dry season, the early wet season and the late wet season. And just frankly speaking, I think Erie really should be exploiting its capacity for multiple generations in the field a lot more. That's just my two cents' worth. Okay, now I'll talk to you about field trials. Again, this is a bit more introduction that a traditional textbook or a pedigree breeding method would look something like this. You have the pedigree line stage of pedigree nurseries where you may have tens of thousands of lines and then you have usually what's called an observational yield trial and then a replicated yield trial followed by that. The OYT is the stage where we first evaluate lines usually after the lines of what we call a fixed. And then after that, we go to a multi-location testing scheme. Just following on what I spoke to you about earlier, it's implying that there were a lot of other alternative schemes. I just thought I'd show you this one. This is a textbook early wheat breeding scheme. It's probably a bit hard to read at the back, but the take home messages are the multi-location trials are a lot earlier than what we would typically do with rice. That's probably the take home message. And I learned a lot of tricks from a lot of breeders back in Australia and this guy in particular, and he had completely different philosophy altogether. He would push it up even earlier, if three or if four actually, but sadly he just passed away last month, but he was quite an innovator in Australian wheat breeding cycles. And again, just to let you know, this is not new information. This was published a long time ago, decades ago. I thought I'd show you this so you can see sort of what's happened to me ever since when I was back in Australia. I've got a lot more gray hair now and I think I've aged terribly on the inside, but I just wanted to, obviously this is not rice. Again, in line with the previous philosophy