Davis, California, USA
DNA LandMarks interview with Dr. Allen Van Deynze, Professional Researcher, University of California, Seed Biotechnology Center.
How long have you been in the field of molecular breeding?
I’ve been doing it since 1989 with RFLPs
In that time what are some of the biggest changes you've seen?
The biggest changes were conversion to PCR markers and now beyond, RAPDs in early 1990s, Simple Sequence repeats in mid 1990s and SNPs in late 1990s. This still made technologies accessible only to major crops. The biggest change in the last 3 years is sequencing technology that makes markers available to any organism, coupled with datapoints in the pennies range.
Given the power of the genomic technology available today where do you see some of the most significant applications of molecular breeding?
As mentioned above, sequencing technology has enabled us to sequence parents and in the near future populations such that we are can truly use genome assisted breeding vs following simple marker associations to major loci. The major change is that we can do denovo marker/quantitative trait loci (QTL) analysis on every population vs just the most important ones. Breeders can have access to the full genotype of plants to combine with breeding value at the phenotypic level. They can also understand the allelic makeup (heterozygous or homozygous at key loci) and begin to understand what genes control complex traits. The current throughput of technologies allows breeders and geneticists to accumulate data on 1000’s of individuals in days to weeks. The challenge is to handle the data.
For a long time, GMO technology has received the most attention (good and bad) in regards to crop biotechnology while genomics has held a lower profile. Is this changing?
The media has amazing power regardless of how good or bad a technology is. Unfortunately, too often the flashy alarming, yet scientifically not credible stories get promoted and published by groups with ulterior motives. Fortunately, this is rarely the case with genomics and it is viewed very positively in media. Who doesn’t think it is cool that a mass murderer can be instantly identified and convicted with a cheek swab, a small fragment of a hair on a crime victim and a good database in less than 1 hour. Programs like CSI do it every week and have given the general public a flavor of the benefits of technology like genomics. Ok it is not that fast, but the general methodology and capability is correct. This is exactly what marker-assisted breeding does in plants---use DNA from plants to create and query large databases of genetic data combined with trait data to decide which plants are likely to give us the best quality, yield, resistances etc.
Given the significant investments by large, multinational ag biotech companies, where do academic institutions such as UC Davis fit into the R&D picture?
Universities are very good at doing research and creating knowledge whereas large multinationals are very good at commercializing products and implementing technologies. UC Davis is pioneering many of the applications and even the genomics technologies themselves. Our strength comes in numbers. We have over 100 plant scientists on campus and have access through the latest technologies through facilities such as our centralized Genome Center. I believe UC Davis plays a central role in developing the biology to understand gene networks and what controls plant traits at the whole plant level across dozens of crop species. Our capacity to analyze the huge volumes of data on comparative genomics – using information from one species to understand others – is equal or larger than most multinationals. Basically, UC Davis and many other public institutions are populating the biological databases across many organisms that enables marker or genome assisted breeding. With over 200 plant commodities in California, we work closely with large and small breeding companies to ensure we are working on the most relevant germplasm and that our research is integrated with the seed industry.
What should medium-sized breeding companies be focusing on to keep pace with their larger competitors?
As any seed company CEO will tell you, germplasm (plant varieties) is the most important asset. Developing the best germplasm in most cost-effective and efficient way is now even more important. There are few crop varieties that last more than 3-5 years in most major crops anymore due to increased investment in plant breeding. Furthermore the first company out with a new trait, such as disease resistance, quality etc gets the lion’s share. Marker assisted breeding can decrease breeding times from 8-10 years to 4-5 years for a variety. Only in the last few years or so has it become cost efficient for medium sized companies to fully use marker assisted selection and whole genome selection on all their crosses vs just the few most important ones. The challenge with marker-assisted selection is that there are dozens of different technologies, with various, cost of reagents, throughput and capital costs. Generally, cost per sample goes down with volume and having a medium number of samples with a medium number of markers is the most expensive. The best systems are flexible and allow one to choose a different set of markers for every cross or population. This is critical as only a small subset of the markers are informative in a given population. These systems are becoming more and more affordable. Another question for a traditional breeding company, is should I invest in my own lab and personnel to run markers, or should out-source it. There are several service companies that have decades of experience in applying markers to breeding. These are a great choice to get started and establish marker assisted programs as the breeding company does not need to make capital and personnel investments and accesses experienced personnel directly. Most often the service companies either already have markers for your favorite crop or can develop them specifically for a breeding company’s germplasm. As the service company must also make a profit, the cost of marker-assisted selection is higher. As a company, one must weigh the cost of the capital investment, depreciation and experienced personnel with the higher cost per datapoint through a service company. In my opinion, marker-assisted breeding is traditional breeding now and is absolutely required to be in business, big or small. Breeders can now select on traits (phenotype) and genotype and make more informed selections.
What will breeding with molecular markers look like 5 years from now? How about 10 years from now?
Hmm, 5 years, It will be integrated in all breeding programs, big and small at some level. We will shift from anonymous markers to those controlling genes. We will be challenged with the huge volumes of data to analyze. In 10 years, we will be using genome (as opposed to marker assisted breeding) assisted breeding and have the sequence of individuals in every cross and have markers that represent the causative mutation for all major traits i.e. diagnostic. As breeders, though, we still won’t be able to predict yield, but be very good at tracking it in every cross. Marker (genome)-assisted selection is a powerful tool in breeding, but, we will always need breeding and breeders and always need to test varieties in the field.
Dr. Van Deynze can be reached at email@example.com