Dieter
K. Mulitze, PhD
President and CEO
Agronomix Software, Inc.
Canada
How can information technology
make seed companies more successful?
Since the mid 1970s, I have been involved in developing software
for plant breeders and have visited plant breeders at seed
companies, universities and government research organizations
around the world. As a former plant breeder myself, I can
readily appreciate the contributions that information technology
can make to plant breeding.
For seed companies and public breeding programs alike,
information technology can increase their success by efficiently
managing and analyzing the plant breeding data and empowering
breeders to make confident, insightful and timely decisions. To
meet this goal, information technology must be easy to use,
powerful, comprehensive, and reliable. A relational database
software system meets those requirements.
Using a single software package to manage variety and pedigree
information, design trials, aid in field operations, and analyze
and store data eliminates the time lost from transferring data
between applications. A relational database permits researchers
to quickly summarize variety performance across locations or
years or instantly find all the data for any genotype, rather
than searching through hundreds of individual files. As a
result, more time and effort can be committed to the more
critical extraction and interpretation of data, thereby
increasing the likelihood of releasing new, royalty-paying
varieties.
Overall, for a software system to increase the success of a seed
company, it must offer at least the following:
(a) |
Data input and organization that is intuitive, efficient, and
flexible. |
(b) |
Support for many practical field operations, such as field
plans or maps for planting thousands of research plots
per site. |
(c) |
Advanced experimental designs and statistical procedures to
control field variation and extract critical information
from research data. Quality decisions require quality
data. Trials lost due to high CVs are costly.
|
(d) |
Timely analysis of research data. |
(e) |
Sharing of data among many researchers via local area networks
(LANs), terminal server, and similar technologies to
enable company-wide decisions in each year or breeding
cycle. |
(f) |
Querying of data to assist decision making, such as extracting
all data on a given genotype or comparing varieties from
all years and locations. This can be critical for GMO
compliance. |
(g) |
Efficient tools to monitor progress in reaching the breeding
objectives. |
(h) |
Flexibility and customization throughout, including the ability
to support many types of breeding schemes for different
crops. |
What specific solutions does your company offer?
We offer AGROBASE Generation II™, a comprehensive and fully
relational software system for plant breeding, agronomy, and
variety testing. AGROBASE Generation II supports agronomic
trials and pedigree data management, many experimental designs
and statistical analysis of data, genetic analyses, genotype x
environment analyses, varietal comparisons, image display of
varieties, and much more, meeting the criteria above. Generation
II runs on stand-alone PCs, on LANs, or on Terminal Server® or
Citrix® for connectivity across a region or globally.
The development of AGROBASE Generation II has capitalized on our
experience in developing an earlier version of the software,
which was adopted by breeders and agronomists in over 70
countries worldwide. Generation II continues to evolve annually,
with improvements and enhancements driven by the needs of our
clients as reflected in currently a new release every six
months. Regional and customized training courses help our
clients to use the software efficiently and realize the maximum
return on their software investment. Saving time, money, and
effort are the goals we set for our software, helping our
clients achieve greater success in the seed industry.
Generation II is licensed as a Basic Agronomic System which is
the foundation or “core” of the software. In addition, there
are four optional modules which may be licensed according to a
client’s needs. The modules are: Advanced Statistics, Varietal
Comparisons, Pedigree Data Management, and Image Display. A new
module will be released late 2007 or early 2008.
What are the more “unique or novel” aspects of your software
solution?
We have found that Microsoft Excel® is widely used, with
breeders and technicians using it on handhelds as well as on
their PCs. Thus, we have developed a direct “express link” which
lets a user export Microsoft Excel® sheets almost instantly from
Generation II, even for an entire location or any logical
grouping of experiments and trials. After data has been entered,
it can be all imported into Generation II seamlessly with
literally a few mouse clicks – even for many thousands of
datapoints. Many of our users appreciate the time saved by this
almost instant “soaking up” of data into a relational database
system.
Seed companies develop new genotypes for many different crops
and plant species, sometimes within the same company. This
involves many breeding schemes – cloning, male sterile lines,
double haploids, hybrid development, synthetics, etc. – readily
supported by the 19 different breeding events available in
Generation II. In addition, the breeder can actually view the
progress across years or generations for progeny from any set of
parents, crosses, or populations. This increases the probability
of success since the breeder should have a way to measure
progress for a given trait and ideally also relate that to any
molecular marker data. With millions of dollars at stake in
developing new genotypes, this is a critical point and goes
beyond simply managing data.
An increasing number of seed companies are using more
sophisticated statistical analyses, especially mixed model
analyses of variety trials or non-orthogonal datasets as well as
various spatial or early generation non-replicated designs.
Rather than “re-invent” the wheel, we have forged a strategic
alliance with VSNi International in the UK, developers of the
world-renowned and widely-used GenStat® statistical package. For
our users who also have a GenStat license and desire even more
advanced analyses, they can now analyze their data directly and
seamlessly from within Generation II while using GenStat. This
has generated a lot of interest, and offers a new “synergy”
between efficient relational data management and the most
advanced of statistical analyses.
With the greater capacity of media storage and the advances in
digital photo storage, more and more seed companies have been
asking for the display of digital images along with their
“normal” numeric and character data. Accordingly, digital photos
can be taken for an experiment, a genotype or variety, a type of
disease or insect reaction, or even for a trait for a treatment
for any given plot in the field or a pot in a greenhouse. Flower
and vegetable breeders in particular appreciate this capability.
Breeders who want to recall what an experiment looked like at
emergence or maturity or harvest several years ago appreciate
the ability to immediately display the digital image.
Agronomix Software, Inc. is a software development company established in 1990 to provide
specialized database management and analysis software for plant
breeders, plant researchers, and agronomists.
For more information about
Agronomix Software, Inc.,
please visit
www.agronomix.com.
Dr. Mulitze can be reached at
mulitze@agronomix.mb.ca
June 2007 |