Source: CORNELL UNIVERSITY submitted to NRP
DSFAS-CIN: EXPANDING THE BREEDING APPLICATION PROGRAMMING INTERFACE (BRAPI) FOR NEXT GENERATION BREEDING METHODS
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
ACTIVE
Funding Source
Reporting Frequency
Annual
Accession No.
1028234
Grant No.
2022-67021-37024
Cumulative Award Amt.
$937,638.00
Proposal No.
2021-11495
Multistate No.
(N/A)
Project Start Date
Jun 15, 2022
Project End Date
Jun 14, 2026
Grant Year
2022
Program Code
[A1541]- Food and Agriculture Cyberinformatics and Tools
Recipient Organization
CORNELL UNIVERSITY
(N/A)
ITHACA,NY 14853
Performing Department
Plant Breeding & Genetics
Non Technical Summary
Modern breeding methods rely on large amounts of phenotypic and genomic data collected across years and environmental conditions. This presents new challenges in effective data management and integration. With recent advances in automation and genotyping methods, the size and complexity of datasets have increased significantly, often resulting in data being stored on multiple systems. Digital agriculture and modern plant breeding approaches increasingly require aggregation of datasets from diverse sources, necessitating data exchange between disparate systems. To facilitate interoperability among breeding applications, the Breeding Application Programming Interface (BrAPI) was developed. BrAPI is a standardized specification for RESTful web service Application Programming Interfaces (APIs) in the plant breeding domain. Initial versions of BrAPI have had a significant impact on data accessibility and usability across major breeding applications. However, the specification needs to continue to grow and evolve to remain relevant as new high-throughput phenotyping and sequencing technologies become mainstream in breeding. Development and maintenance of such a standard is recognized as critical to several important large breeding system initiatives as a foundational technology. This proposal intends to build upon the foundations of the BrAPI standard and continue to develop the BrAPI project. To achieve this, the project will expand the functionality of the RESTful API standard and implement additional API technologies, while maintaining the project infrastructure.
Animal Health Component
50%
Research Effort Categories
Basic
10%
Applied
50%
Developmental
40%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
2017410208060%
2012499108140%
Goals / Objectives
The overall goal of the BrAPI project is to provide the API standards needed to integrate heterogeneous data from multiple data sources to enable advanced analytics and support modern breeding approaches. To achieve this, the proposed project will focus on the following goals:Goal 1: Continue to expand the functionality in the RESTful API standard.Objective 1.1: The BrAPI standard will continue to grow and evolve to suit the needs of the community. The GitHub Issues Board will be used to track issues and minor enhancements, requested and discussed by the community. An iterative approach will refine the standard over time until it can become a best-fit solution for the community.Objective 1.2: New domains will be assessed, developed, and added to the spec, including areas like general analysis, agronomy, crop modeling, phenomics, and sensor data. Expanding the standard to support new domains will allow the community to grow as the standard becomes relevant and useful to more groups.Goal 2: Use the existing data model in other types of API technology.Objective 2.1: The BrAPI data model will be defined and well documented outside the RESTful specification. This will involve extracting all of the JSON model definitions from the RESTful documentation, and creating a schema document to define the models independently. The JSON Schema standard will likely be used to achieve this, and the schema documents will be available to the public through GitHub and the BrAPI website. Extracting the data model into an independent schema will enable easy transfer to any other type of technology.Objective 2.2: The independent data model will be adapted to use in other types of API systems to accommodate more use cases. This requires new documentation, test frameworks, and example implementations to be built for each desired API technology. When possible, automated scripts will be used to create these new digital resources, simplifying the workload.Goal 3: Maintain and expand the existing project infrastructure.Objective 3.1: All existing BrAPI infrastructure will be maintained, including web pages, documentation, and developer tools. With each new version of the BrAPI specification, each of these digital resources will be updated to support the latest version, while still maintaining backwards compatibility with older versions. Some of this maintenance can be automated, but there will always be some amount of manual intervention necessary.Objective 3.2: New documentation, libraries, and testing tools will be developed to support a wide range of new and experienced users. Documentation such as "How To" guides, video tutorials, and interactive training material should allow developers to quickly get up to speed, even if they are new to BrAPI. Language specific libraries and testing tools are very useful to experienced BrAPI developers, providing confidence in working code and sharing maintenance costs across the community.Goal 4: Provide opportunities for direct collaboration between groups.Objective 4.1: Every year, at least two hackathons and a general workshop will be organized. One hackathon will be virtual and one will be in-person (pending the possibility of international travel). Two stipends for travel to in-person hackathons will be available each year. To select travel stipend recipients, candidates will be required to submit a short project description, which will be prioritized by community need and impact. This ranking and selection process will be a joint effort between the Advisory Board and the BrAPI Coordinator.Objective 4.2: Two sabbatical opportunities will be provided each year. To select sabbatical recipients, a similar process will be followed as above. Project descriptions will be submitted to, ranked, and selected by the Advisory Board and the BrAPI Coordinator. Each sabbatical will conclude with a short demo and report of what was accomplished. These reports will be used to determine the effectiveness of the sabbatical opportunities, and allow for adjustments to the process if required.Goal 5: Actively expand the BrAPI community, with a focus on continued diversity.Objective 5.1: Relevant conferences will be attended throughout each year by the BrAPI coordinator, or a suitable representative of the BrAPI community. When possible, the BrAPI coordinator will present information about BrAPI at these conferences.Objective 5.2: News updates and success stories will be generated by the BrAPI coordinator and shared with relevant distribution networks, for example the CGIAR and ELIXIR mailing lists.Objective 5.3: Special training sessions will be organized to introduce new groups to BrAPI, explain what it is and how it works. Because of these efforts, the overall number of partner organizations and compliant software should continue to increase.Achieving these goals will ensure that the BrAPI specification remains relevant, and the community continues to be active and engaged. The standard and the community will grow and adapt to the continually changing landscape of technologies and methods driving modern breeding research.
Project Methods
Goal 1: Continue to expand the functionality in the RESTful API standardThe process for adding new domains to the BrAPI specification is well established. First, a collection of practical use cases will be gathered to represent the requirements of the new API endpoints. Next, data models will be created based on the requirements of those use cases. If there are any existing data standards in the community, those will be considered when creating the data models. As soon as possible, real API implementations will be developed by volunteer early adopters. Finally, the new API will be integrated into the next version of the BrAPI specification and released to the wider community.The new API will be open for comment from the community. The GitHub Issues Board has been adopted for managing problems and minor enhancements to the specification document. This creates a "to do" list of fixes needed for each minor release of the specification, and the input from the community helps guide the priority of each fix. The specification is being developed using an iterative approach, refining the standard with every iteration until it can represent a best-fit solution for most users and use cases. The community has settled on an annual release schedule, combined with a strong enforcement of backward compatibility.Goal 2: Use the existing data model in other types of API technologyMembers of the community have asked for web sockets, message queues, and GraphQL systems with the BrAPI data model. To satisfy this request, the BrAPI data model will need to be extracted from the REST specific documentation. Then, new documentation and example implementations will need to be produced for each of these alternative technologies.Web Sockets, and other Server-Side Event technologies, allow information to be pushed from client to server and server to client. In comparison, REST always makes requests in one direction, from client to server. While REST clients only receive information when they ask for it, Web Socket clients might receive information at any time, provided by the server whenever it is available. This means Web Sockets are more efficient at handling use cases where the server may not have all the data available when a client asks for it but will send it as soon as it is available.Message queues and other message flow technologies typically use a subscription-based connection model. This easily facilitates a one-to-many type of connection between systems, compared to REST, which only supports one-to-one connections. This makes message queues good at broadcasting data to many different systems or collecting data from multiple sensors simultaneously. Often RESTful web services and message queues are used together to solve different data flow problems, so it would be very helpful to have a smooth transition from a BrAPI endpoint to a message queue using the BrAPI data model.GraphQL is an API query framework that provides an enormous amount of flexibility to the client application. Where a RESTful API will define a set list of search parameters the client can use, GraphQL allows a client to build complex queries based on every field in the data model. Creating a BrAPI compatible GraphQL system would allow for enhanced searching for those who needed it, while still using the compatible data models simpler systems will recognize.Goal 3: Maintain and expand the existing project infrastructureThe project website, brapi.org, is the starting point for all BrAPI related inquiries, and requires regular maintenance with updated news, events, documentation, tools, and projects. There are several existing forms of documentation describing BrAPI, including the Apiary and SwaggerHub documentation pages and the project Wiki. All of these pages need regular maintenance to stay accurate and up to date with each released version of the BrAPI standard. The community has been requesting additional forms of documentation, including a formal term dictionary/ontology, a data structure UML diagram, video tutorials, and interactive training materials.The BrAPI website also hosts several tools to assist developers when building BrAPI compliant systems. Currently, there are libraries available to support client applications written in Java, JavaScript, R, and PHP. There are also server stubs packages available in Java and Python. All of these existing libraries require maintenance to stay up to date with the latest version of BrAPI. As the community grows, new libraries will need to be generated and maintained to support a diverse set of programming languages.Testing tools are also available through brapi.org. To test client implementations, the BrAPI Test Server is available with a complete implementation of the standard and dummy data to play with. To test server implementations, the BrAPI Validator (BRAVA) tool is available to confirm the response structure from any BrAPI compatible server. These tools need to be maintained with each new version of the BrAPI specification. Other API technologies described in Objective #2 will require the development of additional testing tools.Goal 4: Provide opportunities for direct collaboration between groupsGiven the importance of the hackathons, the goal of this objective is to organize and provide support for a mix of in-person and virtual hackathons. While the expectation is that most hackathon attendees will pay for their own travel, the BrAPI project would like to support travel for attendees who would not be able to participate otherwise.In addition to hackathons, individual sabbaticals will be supported by funding travel for work sessions. These sabbaticals will allow an individual from one community group to collaborate in person with another group on a specific problem. This will strengthen collaboration between groups and build shared projects, solving use cases for multiple community members simultaneously. A short demo presentation and report will be required at the end of each sabbatical, with some discussion on future work and maintenance of the sabbatical outputs.BrAPI virtual and in-person consulting services from the BrAPI coordinator will continue to be offered by request. Workshops focusing on specific BrAPI topics, including introduction to BrAPI, BrAPI authentication, and deep dives into specific BrAPI modules, will be given at large international meetings. Workshops, meetings, sabbatical opportunities and hackathons will be announced on brapi.org, the BrAPI slack channel, Twitter, and the BrAPI mailing list.Goal 5: Actively expand the BrAPI community, with a focus on continued diversityThe BrAPI project would like to attract the attention of more private breeding and software companies. Many breeding programs operate using proprietary software available to the public by paying licensing fees. Engaging the private sector in the development and implementation of BrAPI can make these proprietary systems more flexible and accessible to users while providing companies the option of leveraging open-source community developed tools. This will provide valuable knowledge and experience to the community, while increasing collaboration between the public and private sectors.Up to this point, BrAPI has been primarily advertised through word-of-mouth, as different research groups begin to collaborate and have a need to share data. This will continue to be an important part of community growth, but it is a passive effort. Active steps can be taken to get the attention of organizations who might benefit from connecting with BrAPI. Participating and speaking at conferences, special training sessions, and press releases for blogs and relevant news outlets are all active steps the BrAPI project can take to advertise.

Progress 06/15/23 to 06/14/24

Outputs
Target Audience: The BrAPI project strives to assist anyone involved with the standardization, transportation, and management of breeding data and related metadata. This includes PI's, breeders, biologists, computer scientists, data scientists, and other experts, working in organizations around the world. The current BrAPI community has about 250 members representing over 37 different organizations involved with breeding research and data management. Changes/Problems:RegardingObjective 1, as discussed in the previous annual report, releasing a new version of the specification every year was found to be unfeasible and undesired by the community. Work has begun on V2.2 which will address the highest priority community needs from the past two years. RegardingObjective 2, the first two tasks are complete as planned. The second two tasks regarding documentation for "Web Sockets" and "Message Queues" will likely get replaced by the need for RDF and NoSQL documentation. These are somewhat newer technologies and will fit the needs of the community much better. RegardingObjective 3, the task "Build an Interactive BrAPI Tutorial" is behind schedule as it was deemed lower priority than some of the other tasks inObjective 3which have proceeded on time as planned. RegardingObjective 4, there was a call for sabbatical project proposals last year, however no one in the community took advantage of the opportunity. The travel funds for the sabbaticals were ultimately reallocated to support travel to the Wittenberg hackathon for those who would not have been able to attend otherwise. Objective 5is proceeding on schedule as planned, with no changes. What opportunities for training and professional development has the project provided?The community hackathons discussed inObjective 4have created opportunities for training and professional development for members of the BrAPI community. Each community hackathon begins with a formal "BrAPI 101" training class to bring everyone up to speed on what BrAPI is, how it works, and how to get started. This class can be requested for specific groups at any time throughout the year, but it will always be presented at least twice a year publicly for the hackathons. This year, there were two additional classes introduced during the hackathons. "BrAPI Domain Model Review" extended the content in the "BrAPI 101" class for more veteran members who wanted to explore a deeper knowledge of the specification. The "Security and Authentication" discussion taught attendees how to use the OAuth2 infrastructure in the context of BrAPI, as well as some general web security best practices. How have the results been disseminated to communities of interest?All news, materials, events, tools and updates in the BrAPI project are communicated to the BrAPI community through the community mailing list, the community Slack workspace, and posted on the project website (brapi.org). The community hackathons are also used as a forum to directly discuss new ideas, gather community feedback, and present important project results. Sometimes, large or important results are communicated to other relevant communities through conference and webinar presentations. What do you plan to do during the next reporting period to accomplish the goals? Develop and release BrAPI v2.2 Enhance the BrAPI Test Server into a production ready server which can be adopted by the community Develop best practice documentation and standards for the BrAPI specification Create a set of self-guided, interactive, training material Organize two hackathons Conduct another round of the BrAPI Sabbatical program Continue attending and presenting at relevant conferences

Impacts
What was accomplished under these goals? Objective 1: Expand the functionality in the RESTful API standard to fully support modern breeding applications. The BrAPI project has established a cycle continuous enhancement and improvement based on the needs of the community. The BrAPI specification version V2.1 was officially released in July of 2022, and since then community groups have been implementing the new standard and sharing feedback. That feedback is collected and tracked through the GitHub issues page, available for public review and commentary. This year, it was determined by the project leadership and advisory board that there are enough issues and enhancement requests to justify the creation of the next version of the BrAPI specification. The development of BrAPI V2.2 officially began in August 2024, but there has been a lot of work done over the past year in preparation for the technical changes. The BrAPI Project Coordinator worked directly with community individuals and teams to discuss each issue. The issues are reviewed, discussed, and expanded on to make sure requirements are properly gathered and documented. One of the main considerations for each change request is to determine if it is specific to a particular team or if it can be generalized and useful to the wider community. This often requires additional discussion with community representatives, gaining general support for an idea during community meetings and hackathons. Objective 2: Implement additional API technologies to achieve optimal performance for a wide range of use cases. During the past year, the BrAPI data model schema has been successfully extracted from the original RESTful documentation. Two community groups were able to successfully use this extracted schema to implement GraphQL API servers. This has been a major goal for the past two years, allowing the BrAPI spec to be used in multiple different types of technologies. Based on this work, the upcoming V2.2 of the spec will be built as a generic schema from the beginning. Then automated tools will be used to create the RESTful documentation, GraphQL documentation, RDF documentation, and any other technologies requested by the community. This change will better support the diversity of the community as well as better future-proof the specification. There are still ongoing discussions in the community about how these different technologies might interact with each other. There is a general consensus to keep REST as the default technology used by most groups, but that may change as different technologies gain favor. Objective 3: Maintain and expand the project infrastructure, supporting software developers to build implementations of the standard quickly and efficiently. Over the past year, most of the existing project infrastructure has remained stable and has required only occasional maintenance. The primary expansion to the infrastructure has been the addition of the automated tools mentioned inObjective 2to manage the BrAPI data model schema. A tool has been built to validate the structure of the schema, making sure it conforms to the JSON Schema protocol, as well as some community best practices specific to the BrAPI project. Development is ongoing for a conversion tool which will be used to take the base schema documents and produce official API documentation (RESTful, GraphQL, RDF, etc). Objective 4: Provide opportunities for direct collaboration between groups, and provide community-building activities. As planned, two BrAPI community hackathons were executed this year. The first, on October 23-27, 2023, was completely virtual. There were 41 participants representing 23 different organizations from around the world, and approximately one fourth of the attendees were new to the BrAPI community. The second hackathon, on April 15-19, 2024, was an in-person event in Wittenberg, Germany, hosted by IPK Gatersleben and FZ Juelich. There were 31 participants representing 19 different organizations from around the world, and more than one third of the attendees were new to the BrAPI community. The biggest collaboration this year was the writing of a new BrAPI Community manuscript. This new paper is intended to be less technical than the first BrAPI paper, with a focus on the different BrAPI compatible tools available. 33 software projects are presented with a short description and an explanation of how BrAPI enhances the tool. This paper was a large collaborative writing project with 76 contributing authors, organized by the BrAPI Project Coordinator. It has been submitted to PLOS Computational Biology, awaiting review. Objective 5: Expand the BrAPI community, with a focus on continued diversity and representation of new domains. Along with the community hackathons, discussed inObjective 4, conferences and webinars are a great way to introduce new people to BrAPI and grow the community. In the past year, the BrAPI project coordinator has presented BrAPI at the following events: the BTI Breedbase Symposium 2023, the Breeding Insight OnRamp Seminar Series, the Cornell Plant Breeding & Genetics Seminar Series, the Plant & Animal Genome Conference 2024, and the NIFA PD Meeting. Each of these events targeted a different community within the domain of international agriculture and breeding. Each event led to follow-up conversations from at least one attendee/group. The company BASF officially joined the BrAPI Community this year. BASF is a major player in private sector breeding and seed production. This new relationship was made possible by some of the work towards a GraphQL standard mentioned inObjective 2. This represents a significant step in one of the BrAPI projects major goals to gain support from more private sector companies.

Publications

  • Type: Conference Papers and Presentations Status: Published Year Published: 2024 Citation: Peter Selby, Applications and Impacts of the BrAPI Project on Plant Breeding, PAG31, January 13, 2024. https://pag.confex.com/pag/31/meetingapp.cgi/Paper/54119


Progress 06/15/22 to 06/14/23

Outputs
Target Audience:The BrAPI project strives to assist anyone involved with the standardization, transportation, and management of breeding data and related metadata. This includes PI's, breeders, biologists, computer scientists, data scientists, and other experts, working in organizations around the world. The current BrAPI community has about 250 members representing over 37 different organizations involved with breeding research and data management. Changes/Problems:In Goal1, the original plan was to release a new version of the specification every year. After polling the community and gathering some anecdotal feedback during the hackathon, the project leadership has decided that this release schedule is too rapid. We need to provide more time for groups to adjust to the new specification before making incremental changes. Also, some of the original targeted development areas have been changed in the project timeline due to community interest and availability. For example, adding "Agronomy and Crop Modeling" concepts was moved up in the timeline as a few members of the community were working on a relevant project they wanted to push a new standard for, while the "Phenomics" concept might be delayed due to funding issues of key community members involved in making that specification. Goal 2 is currently behind schedule according to the original timeline. However, it seems clear that much of the work for this objective is front loaded. Once the first two tasks are complete, the remaining two tasks should be completed quickly. The task "Build an Interactive BrAPI Tutorial", from the original Objective 3, is starting a few months behind schedule as it was deemed lower priority than some of the other tasks in Objective 3 which have proceeded on time as planned. What opportunities for training and professional development has the project provided?Both the community hackathons and the new sabbatical program, discussed in Objective 4, have created opportunities for training and professional development for members of the BrAPI community. Each community hackathon begins with a formal "BrAPI 101" training class to bring everyone up to speed on what BrAPI is, how it works, and how to get started. This class can be requested for specific groups at any time throughout the year, but it will always be presented at least twice a year publicly for the hackathons. The hackathons and sabbaticals both provide opportunities to work with different groups who have different skills and specialties. This often provides informal training as different tools, ideas, and techniques are shared between groups. To formalize this further, one of the sabbatical teams this year had training built into their project goals. Along with the technical work, the traveler wanted to learn an application well enough that he could bring that knowledge back and train the rest of his organization. During the end of sabbatical demo, it was clear that he understood the tool very well and had made technical improvement to it. How have the results been disseminated to communities of interest?All news, materials, events, tools and updates in the BrAPI project are communicated to the BrAPI community through the community mailing list, the community Slack workspace, and posted on the project website (brapi.org). The community hackathons are also used as a forum to directly discuss new ideas, gather community feedback, and present important project results. Sometimes, large or important results are communicated to other relevant communities through conference and webinar presentations. What do you plan to do during the next reporting period to accomplish the goals? Complete separation of BrAPI Data Model from the RESTful documentation Develop a standard GraphQL BrAPI implementation Develop best practice documentation and standards for the BrAPI specification Create a set of self-guided, interactive, training material Write and submit a new manuscript outlining the updates in the BrAPI project since 2018 Organize two hackathons Conduct another round of the BrAPI Sabbatical program Continue attending and presenting at relevant conferences

Impacts
What was accomplished under these goals? Goal 1: The BrAPI specification version V2.1 was officially released in July of 2022. BrAPI V2.1 addresses over 50 problems and enhancements to the specification requested by the BrAPI community. Community requests are tracked through the GitHub issues page and prioritized by general interest and comments left on each issue by members of the community. Issues can vary in size and complexity, ranging from simple spelling errors to the addition of new concepts into the specification. V2.1 introduced 22 new endpoints and 4 new concepts to the specification. These concepts include new data models for pedigree data, sample plate data, agronomic event data, and efficient allele matrix data. Two new working groups were established for looking into specific enhancements for the next version of the BrAPI specification. The first working group is focused on a GraphQL implementation of BrAPI. The second working group is focused on general analysis processes and how to connect with the existing BrAPI specification. This work will make it into the next version BrAPI as a new set of data models and best practice documentation. Goal 2: A new BrAPI community working group was established, focused on building a standard GraphQL implementation of BrAPI. This group has met regularly for the past six months to determine the changes needed to accommodate a GraphQL implementation. The first major task the group identified and began working on involves the extraction of the BrAPI data model schema from the RESTful API documentation. An independent data model schema will define the BrAPI data standard outside of the context of a RESTful API. The standard data models can then be used easily in many other contexts including a GraphQL API system, a database implementation, a message queue system, or any other context where breeding data needs to be represented in transit or at rest. Goal 3: Following the release of BrAPI V2.1, some of the key pieces of the BrAPI project infrastructure needed to be upgraded to match the new specification. The BrAPI Test Server is a public example implementation of the standard with test data allowing the community to test their BrAPI compatible client applications. The Test Server was upgraded to be compatible with V2.1, while maintaining compatibility with all previous versions of the specification. The BRAVA testing tool is an application which can be used to test any BrAPI compatible server. BRAVA was also upgraded to support V2.1, and BRAVA and the Test Server were used against each other to confirm full testing compliance. A number of bug fixes and small enhancements were made to both applications. When both of these tools were fully upgraded, the community could begin their own development efforts with full testing support. During Q1 2023, a number of updates were made to the community website brapi.org. Beyond general maintenance and content updates, three new tools were introduced to help the community. The BrAPI Servers page allows groups to register their BrAPI compatible servers and data sources to be advertised on the website. This tool makes it easy to find other data sources within the community to collaborate with, share data, and test functionality. Similarly, the BrAPI Compatible Software page was also introduced. This page allows community members to register and promote software that is BrAPI compatible, making it easier to share tools and applications with other members of the community. These two pages clearly show the success of the BrAPI project so far by advertising how the community is able to share data, computational resources, and tools using the BrAPI specification. The third tool introduced was the BrAPI Glossary. This is a complete list of all terms used in BrAPI, their definitions, and links to related terms in other standards and ontologies. More enhancements are expected for this tool to better connect it to the independent BrAPI data model described inObjective 2. In addition to these new tools, there were two new informational pages added to the website. The BrAPI Sabbaticals page documents the new sabbatical program (discussed more inObjective 4). This page explains the general information about the program, rules and expectations for applying, and templates to get started with the documentation. It also includes all the documentation and demo recordings from past sabbatical projects. The BrAPI Governance page documents the official governance of the BrAPI project. This includes information on the day-to-day project management team, the community advisory board, and some basic bylaws for nominating and electing advisory board members. This formal governance documentation represents a key piece of the management plan described in the proposal for this project. Goal 4: Two BrAPI community hackathons were executed this year. The first, on October 24-28 2022, was completely virtual. There were 45 participants representing 20 different organizations from around the world, and approximately one third of the attendees were new to the BrAPI community. Many of the projects were focused on upgrading existing implementations. There was a lot of collaboration between groups as they figured out the details of the new specification, with a particular focus on the new pedigree and allele matrix concepts. The second hackathon, was a mixed virtual and in-person event hosted by Clemson University. This was the first in-person BrAPI hackathon in over three years. This was a smaller event than past in-person hackathons, with 20 attendees. Approximately one quarter of the attendees were new to BrAPI. The lower attendance rate was expected as the world continues to recover from the stress of the pandemic. We hope to see more of the community in-person at the planned spring 2024 hackathon. The new BrAPI Sabbaticals Program was established. The sabbaticals are longer form collaboration efforts between groups, where a developer travels to a different organization for up to 6 weeks and engages in an in-person development project. As planned, this new program was able to fund the travel for two teams, one traveler from the US working with a team in France, and another going from Kenya to work with a team in Australia. The project teams were very successful, both with their technical achievements and their international collaboration and team building achievements. The teams have plans to continue working on their projects and maintain the collaborative effort. The final demo for both projects was presented virtually in June 2023, and the call for the next round of project proposals is now open. We feel confident in our ability to repeat the success of this new program next year. Goal 5: Along with the community hackathons, discussed inObjective 4, conferences and webinars are a great way to introduce new people to BrAPI and grow the community. In the past year, the BrAPI project coordinator has presented BrAPI at the following events: the EMBL-EBI 2023 Plant Webinar Series, the AGBT Ag Conference, the NAPDC conference, and the AgBioData Monthly Webinar. Each of these events targeted a different community within the domain of international agriculture and breeding. Each event led to follow-up conversations from at least one attendee/group. This year BrAPI was provided letters of support/collaboration for two new grant proposals regarding high dimensional omics data and genebank ordering systems, respectively. Additionally, BrAPI joined the group of an existing grant (Dsfas-Cin: Integrating Multiscale Remote Sensing Data for Enhancing Data-Driven Predictive Analytics in Crop Breeding and Management) to standardize drone imaging support. If successful, all three of these efforts will increase the size and diversity of the BrAPI community while adding new functionality to the specification.

Publications