High-throughput phenotyping and deep learning to analyze dynamic panicle growth and dissect the genetic architecture of yield formation

Abstract

The dynamic growth of shoots and panicles determines the final agronomic traits and yield. However, it is difficult to quantify such dynamics manually for large populations. In this study, based on the high-throughput rice automatic phenotyping platform and deep learning, we developed a novel image analysis pipeline (Panicle-iAnalyzer) to extract image-based traits (i-traits) including 52 panicle and 35 shoot i-traits and tested the system using a recombinant inbred line population derived from a cross between Zhenshan 97 and Minghui 63. At the maturity stage, image recognition using a deep learning network (SegFormer) was applied to separate the panicles from the shoot in the image. Eventually, with these obtained i-traits, the yield could be well predicted, and the R² was 0.862. Quantitative trait loci (QTL) mapping was performed using an extra-high density single nucleotide polymorphism (SNP) bin map. A total of 3,586 time-specific QTLs were identified for the traits and parameters at various time points. Many of the QTLs were repeatedly detected at different time points. We identified the presence of cloned genes, such as TAC1, Ghd7.1, Ghd7, and Hd1, at QTL hotspots and evaluated the magnitude of their effects at different developmental stages. Additionally, this study identified numerous new QTL loci worthy of further investigation.

Keywords

Deep learning / Growth prediction / High-throughput phenotyping / Panicle growth / QTL / Rice

References

[1] Bai S., Hong J., Li L., Su S., Li Z., Wang W., Zhang F., Liang W., Zhang D., 2021. Dissection of the genetic basis of rice panicle architecture using a genome-wide association study. Rice 14, 77.
[2] Busemeyer L., Ruckelshausen A., Möller, K., Melchinger A.E., Alheit K.V., Maurer H.P., Hahn V., Weissmann E.A., Reif J.C., Würschum T., 2013. Precision phenotyping of biomass accumulation in triticale reveals temporal genetic patterns of regulation. Sci. Rep. 3, 2442.
[3] Chen S.H., Zhou L.J., Xu P., Xue H.W., 2018. SPOC domain-containing protein Leaf inclination3 interacts with LIP1 to regulate rice leaf inclination through auxin signaling. PLoS Genet. 14, e1007829.
[4] Desai S.V., Balasubramanian V.N., Fukatsu T., Ninomiya S., Guo W., 2019. Automatic estimation of heading date of paddy rice using deep learning. Plant Methods 15, 76.
[5] Guo W., Fukatsu T., Ninomiya S., 2015. Automated characterization of flowering dynamics in rice using field-acquired time-series RGB images. Plant Methods 11, 7.
[6] Jung K.H., An G., Ronald P.C., 2008. Towards a better bowl of rice: assigning function to tens of thousands of rice genes. Nat. Rev. Genet. 9, 91-101.
[7] Kang K., Park S., Natsagdorj U., Kim Y.S., Back K., 2011. Methanol is an endogenous elicitor molecule for the synthesis of tryptophan and tryptophan-derived secondary metabolites upon senescence of detached rice leaves. Plant J. 66, 247-257.
[8] Li B., Chen L., Sun W., Wu D., Wang M., Yu Y., Chen G., Yang W., Lin Z., Zhang X., Duan L., Yang X., 2020a. Phenomics-based GWAS analysis reveals the genetic architecture for drought resistance in cotton. Plant Biotechnol. J. 18, 2533-2544.
[9] Li H., Feng H., Guo C., Yang S., Huang W., Xiong X., Liu J., Chen G., Liu Q., Xiong L., Liu K., Yang W., 2020b. High-throughput phenotyping accelerates the dissection of the dynamic genetic architecture of plant growth and yield improvement in rapeseed. Plant Biotechnol. J. 18, 2345-2353.
[10] Li Z., Sillanpää, M.J., 2015. Dynamic quantitative trait locus analysis of plant phenomic data. Trends Plant Sci. 20, 822-833.
[11] Liu T., Liu H., Zhang H., Xing Y., 2013. Validation and characterization of Ghd7.1, a major quantitative trait locus with pleiotropic effects on spikelets per panicle, plant height, and heading date in rice (Oryza sativa L.). J. Integr. Plant Biol. 55, 917-927.
[12] Lu H., Cao Z., Xiao Y., Fang Z., Zhu Y., Xian K., 2015. Fine-grained maize tassel trait characterization with multi-view representations. Comput. Electron. Agric. 118, 143-158.
[13] Sandhu J., Zhu F., Paul P., Gao T., Dhatt B.K., Ge Y., Staswick P., Yu H., Walia H., 2019. PI-Plat: a high-resolution image-based 3D reconstruction method to estimate growth dynamics of rice inflorescence traits. Plant Methods 15, 162.
[14] Song J.M., Xie W.Z., Wang S., Guo Y.X., Koo D.H., Kudrna D., Gong C., Huang Y., Feng J.W., Zhang W., Zhou Y., Zuccolo A., Long E., Lee S., Talag J., Zhou R., Zhu X.T., Yuan D., Udall J., Xie W., Wing R.A., Zhang Q., Poland J., Zhang J., Chen L.L., 2021. Two gap-free reference genomes and a global view of the centromere architecture in rice. Mol. Plant 14, 1757-1767.
[15] Tang J., Zhu X., Wang Y., Liu L., Xu B., Li F., Fang J., Chu C., 2011. Semi-dominant mutations in the CC-NB-LRR-type R gene, NLS1, lead to constitutive activation of defense responses in rice. Plant J. 66, 996-1007.
[16] Wang J., Yu H., Xie W., Xing Y., Yu S., Xu C., Li X., Xiao J., Zhang Q., 2010. A global analysis of QTLs for expression variations in rice shoots at the early seedling stage. Plant J. 63, 1063-1074.
[17] Wang X., Zhou T., Li G., Yao W., Hu W., Wei X., Che J., Yang H., Shao L., Hua J., Li X., Xiao J., Xing Y., Ouyang Y., Zhang Q., 2022. A Ghd7-centered regulatory network provides a mechanistic approximation to optimal heterosis in an elite rice hybrid. Plant J. 112, 68-83.
[18] Wu W., Liu T., Zhou P., Yang T., Li C., Zhong X., Sun C., Liu S., Guo W., 2019. Image analysis-based recognition and quantification of grain number per panicle in rice. Plant Methods 15, 122.
[19] Wu X., Feng H., Wu D., Yan S., Zhang P., Wang W., Zhang J., Ye J., Dai G., Fan Y., Li W., Song B., Geng Z., Yang W., Chen G., Qin F., Terzaghi W., Stitzer M., Li L., Xiong L., Yan J., Buckler E., Yang W., Dai M., 2021. Using high-throughput multiple optical phenotyping to decipher the genetic architecture of maize drought tolerance. Genome Biol. 22, 185.
[20] Xie E., Wang W., Yu Z., Anandkumar A., Alvarez J.M., Luo P., 2021. SegFormer: Simple and efficient design for semantic segmentation with transformers. Adv. Neural Inf. Process Syst. 34, 12077-12090.
[21] Xie W., Feng Q., Yu H., Huang X., Zhao Q., Xing Y., Yu S., Han B., Zhang Q., 2010. Parent-independent genotyping for constructing an ultrahigh-density linkage map based on population sequencing. Proc. Natl. Acad. Sci. U. S. A. 107, 10578-10583.
[22] Xing Y., Zhang Q., 2010. Genetic and molecular bases of rice yield. Annu. Rev. Plant Biol. 61, 421-442.
[23] Xiong X., Duan L., Liu L., Tu H., Yang P., Wu D., Chen G., Xiong L., Yang W., Liu Q., 2017. Panicle-SEG: a robust image segmentation method for rice panicles in the field based on deep learning and superpixel optimization. Plant Methods 13, 104.
[24] Yang W., Guo Z., Huang C., Duan L., Chen G., Jiang N., Fang W., Feng H., Xie W., Lian X., Wang G., Luo Q., Zhang Q., Liu Q., Xiong L., 2014. Combining highthroughput phenotyping and genome-wide association studies to reveal natural genetic variation in rice. Nat. Commun. 5, 1-9.
[25] Yu B., Lin Z., Li H., Li X., Li J., Wang Y., Zhang X., Zhu Z., Zhai W., Wang X., Xie D., Sun C., 2007. TAC1, a major quantitative trait locus controlling tiller angle in rice. Plant J. 52, 891-898.
[26] Yu H., Xie W., Wang J., Xing Y., Xu C., Li X., Xiao J., Zhang Q., 2011. Gains in QTL detection using an ultra-high density SNP map based on population sequencing relative to traditional RFLP/SSR markers. PLoS One 6, e17595.
[27] Yu Y., Ouyang Y., Yao W., 2018. shinyCircos: an R/Shiny application for interactive creation of Circos plot. Bioinformatics 34, 1229-1231.
[28] Zhang Q.,2007. Strategies for developing green super rice. Proc. Natl. Acad. Sci. U. S. A. 104, 16402-16409.
[29] Zhou Y., Srinivasan S., Mirnezami S.V., Kusmec A., Fu Q., Attigala L., Fernandez M., Ganapathysubramanian B., Schnable P.S., 2019. Semiautomated feature extraction from RGB images for sorghum panicle architecture GWAS. Plant Physiol. 179, 24-37.
[30] Zhu Z., Tan L., Fu Y., Liu F., Cai H., Xie D., Wu F., Wu J., Matsumoto T., Sun C., 2013. Genetic control of inflorescence architecture during rice domestication. Nat. Commun. 4, 2200.

Funding

* E-mail address: [e-mail] (W. Yang).