Development and application of a Data Management and Acquisition System for pear breeding

Abstract:【Objective】Fruit tree breeders require a flexible and efficient method to collect breeding data in the field and to integrate other different kinds of breeding data. This paper introduced a pear breeding data management and acquisition system, which included a PC application performed as the data center and an Android application for data collection in the field. This system would improve the efficiency of field work in pear breeding.【Methods】The system was designed to support bidirectional control and data transmission between PC(data center) and the Android application via a 2- dimensional(2D) barcode identification(ID) tag hanged on the pear tree. In order to achieve this goal, different combinations of tag materials, encapsulation methods and hanging modes were screened continuously in fields from 2010 to2014 at Yangdu research station of ZAAS in Zhejiang province. The tags were produced by Agrox?printers and hung on 288 pear trees(from two F1 progenies,‘Cuiguan'בQingxiang'&‘Qingxiang'בCuiguan', planted in 2008). The unit cost and corrosion resistance performance of tags were compared with the traditional marking ways(painting on tree stem or using plastic tags). The readability of QR code were also examined. The PC application was developed in C# using.NET framework(Version 2.0) in MicrosoftVisual Studio(Version 2010) based on Microsoft Access(Version 2008) database. The Android application(App) was built in Action Script?(Version 3.0) in Adobe Flash builder?(Version 4.6) with SQLite?(Version 3.6) as the database. The App was validated for consistency and stability on the handheld device“i Data?95”. To evaluate the data acquisition efficiency of the App compared with the traditional manual methods, a test was carried out in lab by three groups of nine graduate students who had no any experiences of pear breeding and never used the system. Before the test, students received 20 minutes' training to be familiar with the handheld device and the App. During the test, the time elapsed for writing the values of sixteen characteristics and taking three pictures of three simulated pear progenies was recorded for each student. Based on these information, we calculated the data collection efficiency between the App and the traditional manual methods【.Results】The optimized material and processing method of QR tags was selected to be better suited to pear progenies, which were made of red plastic coated paper(50 mm×90 mm) with barcode printed by a thermal printer and hanged on trees with U-shaped nail in the same direction, after encapsulation with plastic package. The tag had been proved to be a long life of more than five years and easy to be found and read QR code in fields. The tag cost was pretty cheap: only about 0.21 RMB Yuan per tag. The system consisted of two subsystems: one was a PC application for(1) storing basic information,such as the address of bases & fields, resources list and resource related documents & pictures, field planting map, etc.;(2) controlling the QR code generation and tag printing;(3) defining the value of personalized observing characteristics(in addition to the system built-in character, which were built based on the Descriptors and Data Standard for Pear(Pyrus spp.) and Guidelines for the Conduct of Tests for Distinctness,Uniformity and Stability-Pear);(4) managing field records and data synchronizations with handheld devices; The other subsystem consisted of two Android Apps, which were used not only for tracking the ID information, but also for collecting and storing various observed values(such as text, picture, voice, geographic coordinates, etc.) of pear germplasm resources & hybrid populations in the field, respectively. Efficiency test showed that four procedures: recording data, taking pictures, logging data in Excel and renaming pictures must be required in manual data acquisition methods, but only the first two procedures were needed by using Android App in handheld device. The average time used by manual recording was 201.1 seconds,while the average time used by handheld device recording was only 155.3 seconds. As a result, the time of data entry could decrease at least 25% by using handheld devices, compared with the traditional data logging methods【.Conclusion】The designed tags had the features of low cost, durable, and easy to be identified in the field, which was suitable for the identification of germplasm resources and progenies of pear and other fruit crops, including apple, peach and orange. The pear breeding data management and acquisition system had many advantages, including centralized management of various breeding related data, the ability to collect data in the field by using the handheld device instead of the traditional manual data collecting method, to import and export the data without internet availability and to use the exported data directly in spreadsheet or statistical analysis software. The system was proved to be an effective tool for improving the pear breeding work and providing a significant productivity boost of breeders.