Jean-Marcel Ribaut
CIAT
20 October 2015, Cali Colombia
Lessons learnt from the GCP
experience and IBP next steps

GCP in Brief
 A CGIAR Challenge Programme hosted at CIMMYT
 10-year framework (Phase I, 2004–2008; Phase II, 2009–2014)
 US$ 170 M program
 Target zones: drought-prone environments
 Sub-Saharan Africa, South & South East Asia, L. America
 Eighteen CGIAR mandate crops in Phase I
 Nine CGIAR mandate crops in Phase II
 Cereals: maize, rice, sorghum, wheat,
 Legumes: beans, chickpea, cowpea, groundnut
 Roots and tubers: cassava
Strategic objective: To use genetic diversity and advanced plant science to
improve crops for greater food security in the developing world
GCP: A broker in plant science bridging the gap between upstream and applied
science
www.generationcp.org

The sorghum case: From Cornell to African farmers’
fields with a stopover in Brazil: a ten-year effort
 Step 1: Competitive Project (initiated 2004)
 Led by Cornell in collaboration with EMBRAPA
 Plantlets screened under hydroponics – Alt1 Gene cloned
Magalhaes et al. 2007, Nature Genetics, 39: 1156–1151
 Step 2: Competitive Project (initiated 2007)
 Led by EMBRAPA in collaboration with Cornell
 Favourable alleles identified – Improved germplasm for
Brazil
Caniato et al. 2011, PLoS One 6, e20830
 Step 3: Commissioned work (initiated 2009)
 Led by Moi University in collaboration with EMBRAPA
 Introgression of favourable alleles – Improved germplasm
for Kenya and Niger
Linking Upstream with Applied Science

Phase II

 Genetic resources
 Reference sets for 18 crops (all CGIAR mandate crops)
 Genomic resources
 Markers for ‘orphan crops’
 Informative markers
 Drought, viruses and insect resistance
 Genes/QTL
 AltSB for aluminium tolerance, Pup1 for P uptake efficiency, Saltol for
salt tolerance and Sub1 for submergence tolerance.
 Improved germplasm
 New bioinformatic tools (DM, diversity studies, breeding, etc)
 Enhanced capacities for MAB in NARS programmes
 Human-resource capacities / Physical infrastructure / Analytical power
 Ex-ante analyses on MB impact in developing countries
Product Catalogue available at:
www.generationcp.org/impact/product-catalogue
Selected major research outputs

‘Classic’ approach
 Formal postgraduate training programmes
 100+ MSc and PhD students whose work is embedded in research projects
 Workshops, fellowship grantees, travel grants
 Train-the-trainers for future regionalised capacity-building sustainability
 Communities of practice
 Rice in the Mekong; Cassava in Africa
 IBP-hosted (both crop- and expertise-based)
Perhaps not so common – probably uniquely GCP
 Capacity building à la carte
 Integrated Breeding Multi-Year Course: breeding, data management,
data analysis
 CB along the delivery chain (scientists, technicians, station managers)
 Technical support for infrastructure implementation
 IBP an integrated way to promote the problem-solving approach
 It is really about “learning as you go”
Capacity Building

Building on our Network!
IB-MYC:134 breeders; 10 crops; 31 countries

Key learning areas
 Governance
 Research management
 Monitoring and evaluation
 Partnership
 Adoption and behaviour change
 Integrated Breeding Platform
 Conclusion

Governance
Issue:
 Dysfunctional governance for nearly half of GCP’s life until
mid-2008, with governance body comprised of direct
beneficiaries of its own decisions
Solution:
 Involvement of stakeholders (‘owners’) and partners to
define the overall objectives and general direction, but
 Separate independent body to approve workplan and
oversee implementation
 Small group of complementary expertise (GCP EB works very well!)
 Access to specific expertise when needed (eg, GCP’s IP Committee)
Accountability must be clarified first!

Science management:
broker in plant science – the CP model
A management team that defines and implements ‒ in
partnership and through grants ‒ a workplan to achieve
overall objectives
Agile research management approach that allows…
 Bringing in new ideas for strong partnerships
 Continually enhancing research quality and efficiency
 Adjusting research activities based on external environment
 New technologies, partners, opportunities for synergy, etc
 Easily discontinue unsuccessful projects
But…
 Must revolve around a specific research topic
 Can only exist with the support of well-established institutes
 Ideally focused and time-bound
 Excellent complement of core activities

 Competitive grants
 Do not necessarily fall neatly into your research priorities (dead-end
projects)
 Capture emerging opportunities, best ideas and new partners
 Increase research quality
 Commissioned projects
 Not always good value for money, less transparent
 Consolidate the research agenda
 Very efficient when they build on successful competitive projects
Different kinds of research: the dynamics
Competitive
Commissioned
Services
10 years
$
Competitive and commissioned approaches each have their pros
and cons, but are complementary, maximising impact & potential!

Monitoring and evaluation
Issue:
 Inadequate research-management capacity in GCP’s early
years due to part-time appointments (attractive in theory, but
difficult in practice)
 Lack of an M&E framework from the beginning (though this may
not have been required at the time)
 Conflict of interest within the MT
 Not the same skills (Animator vs Manager)
Options:
 Full-time management team
 Separate the planning and implementation from a stand-alone
M&E component
Of course good management capacity and practice have a
cost, and therefore efficiency needs to be considered
carefully
(Management cost of the GCP: about 15%)

Quality of the Science:
Peer-reviewed publications
5
25
51
57
68
78
73
90
32
0
10
20
30
40
50
60
70
80
90
100
2005 2006 2007 2008 2009 2010 2011 2012 2013
Journal articles published: 2005‒2013
Year
Number
In selected high-impact journals (2007–2013):
Nature – 5; Nature Biotech – 3 Nature Genetics – 2; PNAS – 8

 Be strategic in partnership development
 The importance of people
 People are first, and institutions are second
 Building on existing partnerships, maximising on personal relations
 Be selective, and also cautious
 Can easily get out of hand, can be a distraction
 Plan for it, and do not underestimate effort needed
 managing true partnerships takes time and resources!!!
 But, if managed well…
 One of the most efficient and effective ways to do business
 One of the most rewarding components of the work
 Creates a special group dynamic and brings in new ideas
 Cultivates public trust, with the resultant positive public image
Partnership: key notes to keep in mind

Indicators
 Money allocation to partners
 Significant in-kind contribution from partners
 Open exchange of experience and information
 Partners not necessarily attracted (purely) by money, but to be part
of a network, visibility and exchanges with peers abroad
 Critical but indispensable intangibles – trust and goodwill
 Partners continue to work together after GCP projects end
Evolution of roles and responsibilities
 A switch: Leaders become mentors
 Knowledge applied & transferred: Trainees become doers & leaders
 In phase II, more than half of our PIs are from developing countries
and more than half the grants go directly to National Programmes
It takes time and resources to nurture and implement true
partnership!
True Partnerships

GCP network
EMBRAPA
Brasilia
Brazil
CIP
Lima
Peru
CIAT
Cali
Colombia
CIMMYT
Mexico City
Mexico
Cornell
University
USA
Wageningen University
Netherlands
John Innes Centre
Norwich
UK
CAAS
Beijing
China
NIAS
Tsukuba
Japan
Agropolis
Montpellier
France
IPGRI
Rome
Italy
WARDA
Bouaké
Cote d’Ivore
IRRI
Los Baños
Philippines
ICRISAT
Patancheru
India
ICARDA
Aleppo
Syria
IITA
Ibadan
Nigeria
ACGT
Pretoria
South Africa
ICAR
New Delhi
India
BIOTEC
Bangkok
Thailand
INRA
Rabat
Morocco
CINVESTAV
Irapuato
Mexico
Instituto Agronomico per l’Oltremare
Florence
Italy
9 CGIAR
6 ARIs
7 NARS
ETH
Zurich
Switzerland
Partners
Consortium

 EPMR panel (2008) noted that the GCP community is one
of the Programme’s most crucial assets. In their words:
“Perhaps the most important value of GCP thus far, is the
opportunities it has provided for people of diverse backgrounds
to think collectively about solutions to complex problems, and,
in the process, to learn from one another.”
Major achievement: the GCP community
GCP People: The Programme’s Greatest Asset!

Other challenges
Operational
 Keeping key partners aligned with the overall shared
objective(s)
 Prioritisation and resource allocation
 The two bosses and part-time boss syndromes
 Communication (internal and external) – vital for a
distributed team
 Recognition and ownership
Research
 inclusiveness vs efficiency
 Germplasm exchange
 Genetic stocks
 Work quality standard
 Inclusiveness vs efficiency

The Integrated Breeding
Platform

Challenges:
 Most of the breeders in the developing world capture their data
by hand and store them in hard copy (book)
 Protective and proprietary attitude prevents data sharing
 Not a top priority, no clear resource allocation, data still in the
hands of individual scientists
One of the major challenges in collaborative efforts (GCP)
Implementation:
 Clear DM policy in place at the institutional level
 Quality and documentation improved thanks to:
 Adoption of new data capture tools with predefined templates
 Proper budget allocation, support staff, part of the staff evaluation process
 Donor requirement beforehand
 Quality control must start at the scientist level
5-20% of crop improvement R4D is wasted because of poor data
management (quality, analysis, availability) and breeding decisions
Data Management (A Key Technical Hitch)

The IBP is a comprehensive web-based platform enabling
plant breeders to access and use modern breeding in a
practical and sustainable way
 We are here to support breeders to access good breeding
practices, enabling them to play a leading role in R4D
 Our primary target are breeders in developing countries, both in
the public and private sectors
 We target sustainable deployment at the institutional level
 We promote the adoption of the entire platform (products,
services and networks)
IBP adoption would represent a small revolution for breeders
The right platform is a must have to maximize genetic gain
To adopt the platform must go with a revision of breeding strategy
Integrated Breeding Platform (IBP):
Definition and core principles

Breeders:
 Increase data quality, documentation and exchange
 Savings in time and cost to run breeding activities
 Increased genetic progress per each crop cycle
 Enhanced certainty in crop breeding outcomes
Value Proposition
Institutions:
 Improved institutional data management
 Better product at a lower price (efficiency and effectiveness)
 Improve the value proposition to attract funding
Society:
 Improved crops (quality and yield) in farmers’ fields
 More income for smallholder farmers
 More and better food to feed the world

 CGIAR Centers:
♦ Ongoing: Africa Rice, CIMMYT, ICARDA, ICRISAT and IITA (CIAT and CIP)
 Selected Institutions:
♦ AATF (Africa), CIRAD (France), ACPFG (Australia), Nobel Foundation (USA)
 Selected contact in the Private sector
♦ Biogemma, Limagrain, Seed Co, Monsanto
 Formal commitment of NARS Directors in Africa and Asia
♦ ARC (Sudan) BIOTEC (Thailand) CAAS (China), EIAR (Ethiopia), IER (Mali),
PhilRice (Philippines), ISRA (Senegal), NARO (Uganda) and Regional
Centers in Tanzania
 Universities in developed countries:
♦ Ongoing: Cornell, UGA, ISU, Riverside (Davis, Hohenheim, IRTA, WUR)
 Universities in developing countries:
♦ Ongoing: KwaZulu-Natal, Makerere, WACCI +11
 Users
♦ About 1500 people download the BMS
♦ About 300-500 early adopters of the BMS worldwide
BMS Deployment and Early Adopters

Tools
 Breeding Management System (BMS)
 The core product of the IBP
 Peripheral tools (Collaborations):
 Electronic data capture devises, code bare, electronic weights, LIMS, etc
Crop Information
 Crop ontology and trait dictionary
Breeding Services
 Genotyping (e.g. SNP sets), diagnostic markers, germplasm
Capacity Development
 Tutorials, courses and online teaching
Community Spaces
 Blogs, forums, publications
IBP Core Products

Breeding Management System
A suite of interconnected software tools and applications specifically
designed to help breeders manage their day-to-day activities:
Programme management
Customise preferences and
monitor programme activities
from the Workbench, a
dashboard application with
integrated tools to manage and
query crop information across
the system
Marker-assisted breeding
Select germplasm and design
crosses by complementing
phenotypic selection with marker
technology, for integrated breeding
decisions
Breeding activities
Prepare trials and nurseries,
manage seed inventories and
keep continuous genealogy
records season after season
Statistical analysis
Analyse field and lab data with
powerful statistics and mixed
model comparisons of locations
and genotypes

Molecular Breeding Pipelines
Three pipelines ready by early 2016
 Seed purity:
 Pre-selected set of markers
 Neutral plus diagnostic ones
 Distribution along the genome
 Just a must have
 Elite allele enrichment
 2-3 must have genes for a breeding programme
 Select H or homozygote favorable alleles
 Early recombination stage (thousands of F2 plants)
 Large economical impact
 Elite allele introgression (MAS, BC-MAS)
 Background recovery
 Confirmation of genetic effect in elite background
 Often major genetic gain for a target trait

https://www.integratedbreeding.net

Deployment matrix

Deployment matrix

Adoption Flowchart – Champion path

 Most people are reluctant or resistant to change
 Even people who are interested in change often do not allocate
the time and resources to effect change
 Even where there are clear and demonstrable benefits from
making a change, this alone is not sufficient incentive
 Most changes can be implemented only by:
 Strong bottom-up demand
 Mandatory top-down decision
 Need to persuade people to be ready to:
 Get out of their comfort zone
 Dedicate time to learning new things, even if that might not benefit their
work directly, or immediately
 Adopt a collaborative rather than competitive approach
 Enforcement and implementation
 Big difference between the private and public sector
Changing people’s behaviour:
A real challenge in technology transfer

Local Support Services:
Network of Hubs

Budget and Business Model
 Phase I (Traditional/Safe):
 2009-14: $20M ($12M for Bill & Melinda Gates Foundation) under the
leadership of the GCP
 “Classic technology development project”
 Phase II (Transition/Challenging):
 2014-19: $24M ($16M secured) under the leadership of the IBP
 2015-2016 a critical period, conservative expenditures
 BMS Commercial version early 2016
 Key Performance Indicators (KPI) for all Staff (starting with the Director)
 Performance Matrix for Regional Hubs
 Phase III + (Non profit Commercial Organization/???)
 2019-who knows: ???, no funding secured, under the leadership of the
GCBSS
 Self-funded for the maintenance and development of the BMS
 Deployment and capacity building from public funds
A great opportunity to develop a new business model in R4D
Building on the ENTREPRENEURIAL spirit

The Business Model!
The Money Pot
Donors
Development Deployment
Commercial
Capacity
Development
Happy subsidized
Users
Happy Paying
Users

Conclusions

To think about…..
 The engagement of our breeders to work very closely with the
champions is paying off
 Trainees become:
 Trainers (knowledge dissemination)
 Mentors (helping others)
 Champions (testimony)
 Ambassadors (promotion)
 BMS deployment must go with modernization of breeding
practices
 The BMS is not supposed to replicate what the breeders are
currently doing but is aimed at implementing a breeding
strategy defined at the programme/Institutional level

Conclusions
 Major achievements have probably revolved around:
 Establishing true partnership with cultural change on how to run R4D
projects
 Several flagship projects
 Dynamic environment learning by doing (Research and management)
 Enabling partners in developing countries to:
 Play a leadership role
 Access modern breeding
 We also had some clear shortcomings
 Monitoring & evaluation were the biggest shortfalls (data management)
 Several competitive projects were dead-ends
 Lessons learnt from the CPs in general and GCP in particular can
inform other operational and organisational models
 IBP will survive GCP
 Cross cutting support platforms are major assets to implement R4D
effectively and efficiently

GCP/IBP International Staff 2003 - 2014
 Akinola Akintunde
 Antonia Okono
 Arllet Portugal
 Carmen de Vicente
 Chunlin He
 Clarissa Pimentel
 Claudia Bedoya
 Corina Habito
 Delphin Fleury
 Diego González-
de-León
 Eloise Phipps
 Fernando Rojas
 Fred Okono
 Gillian Summers
 Graham McLaren
 Hamer Pascal
 Hei Leung
 Humberto Gomez
 Jan Erik Backlund
 Jean Christophe
Glaszmann
 Jenny Nelson
 Jonathan Crouch
 Kaitlin Lesnick
 Kate Durbin
 Larry Butler
 Mae Christine
Maghirang
 Maria Teresa Ulat
 Mark Sawkins
 Ndeye Ndack Diop
 Nelzo Ereful
 Nosisa Mayaba
 Peter Ninnes
 Philippe
Monneveux
 Rajeev Varshney
 Robert S. Zeigler
 Rowena Tulod
 Shawn Yarnes
 Theo van Hintum
 Valérie Boire
 Xavier Delannay