Collins Udanor – University of Nigeria Nsukka - Nigeria (collins.udanor@unn.ng)

1. Development of a SGW-based Plant Tissue Culture Micropropagation Yield Forecasting Application, Plantisc2 Collins Udanor – University of Nigeria Nsukka - Nigeria (collins.udanor@unn.edu.ng) WACREN e-Research Hackfest – Lagos (Nigeria)

2. Outline • Scientific problem area • Starting point • Technology stack • Computational and data model • Implementation strategy

3. Scientific problem • Abstract: Plant tissue culture is a collection of techniques used to maintain or grow plant cells, tissues or organs under sterile conditions on a nutrient culture medium of known composition. Plant tissue culture is widely used to produce clones of a plant in a method known as micropropagation. During the UNESCO-HP Brain Gain Initiative (BGI) project (2009-2013), the University of Nigeria team conducted series of plant tissue culture experiments and developed a stand-alone application, Plantisc. A Plant Tissue Culture micro propagation simulation software, which achieved over 67% predication accuracy whose result was published in a peer-reviewed journal . [http://elvedit.com/journals/IJACSIT/wp-content/uploads/2014/06/Gridsim-Paper-2.pdf]

4. What scientific domain is this application addressing ? Plant Science and Biotechnology • What are the identified problems that this application tries to solve ? • Time taken to perform the experiment is much • the experiment is cost intensive • and it is still in an empirical stage. • What do you see as some of the benefits of using the web for this application? • Availability of the application to a wider research community • Reduction in cost of experiment • Reduction in time of experiment

5. Fig 1. Tissue Culture experiments in the Lab

6. Workflow • The various workflows that users of the application will undertake will include: • The application will have a user input GUI for the user to upload his/her data • The data is read from a .csv file or other format and stored as an array in a Python function • The user selects what part of the plant tissue he/she wants to forecast the yield, e.g. root, shoot, leaf • The user selects different combinations of auxins at various quantities • The user submits the job • The user checks the status of the job • The user retrieves the result when completed • The user analyzes the results • The cycle is repeated as the need may arise.

7. Figure 2: Input capture form

8. Figure 3: Output

9. Data model • Please describe : • The various workflows that users of the application will undertake Described in the last slide (4) • Data origins, ingestion, management : From experiments in Plant Tissue culture labs • Where are you getting data from ? Data can be stored in repositories anywhere, PC or cloud repos • Where should it be moved to ? Data may be moved to application database or cached during runtime • Where should it finally be stored ? In a cloud repository for a specific number of days

10. Computing model • Which of these models apply to you ? • High throughput computing (grid/cloud/p2p) • High-performance computing (grid/HPC) The above types

11. Implementation strategy • We need you to develop a project plan • Strategy: • - Develop Regression models • - Modular development • - Unit testing • - Share on Github • Mid-hack checkpoint - Should have Created user interface - Upload user data - Implemented Regression model • End-of hack checkpoint Full functional application Deployable on the cloud - Upload user data by mid December - Implemented Regression model - by end of December - Full functional application - By mid January - Deployable on the cloud - By end of January 2017

12. Risks and unknowns • What do you think will get in your way ? List anything, and try to estimate the negative impact that it will have Envisaged challenges may include: • Time factor - Official work hours may slow the speed of development • Lack of support from other programmers - most of my programmers are outside the country on study

13. Thank you! sci-gaia.eu info@sci-gaia.eu