A Deep Learning-Based Phenotypic Analysis of Rice Root Distribution from Field Images

Root distribution in the soil determines plants' nutrient and water uptake capacity. Therefore, root distribution is one of the most important factors in crop production. The trench profile method is used to observe the root distribution underground by making a rectangular hole close to the crop, providing informative images of the root distribution compared to other root phenotyping methods. However, much effort is required to segment the root area for quantification. In this study, we present a promising approach employing a convolutional neural network for root segmentation in trench profile images. We defined two parameters, Depth50 and Width50, representing the vertical and horizontal centroid of root distribution, respectively. Quantified parameters for root distribution in rice (Oryza sativa L.) predicted by the trained model were highly correlated with parameters calculated by manual tracing. These results indicated that this approach is useful for rapid quantification of the root distribution from the trench profile images. Using the trained model, we quantified the root distribution parameters among 60 rice accessions, revealing the phenotypic diversity of root distributions. We conclude that employing the trench profile method and a convolutional neural network is reliable for root phenotyping and it will furthermore facilitate the study of crop roots in the field.

Accumulation of additive effects generates a strong photoperiod sensitivity in the extremely late-heading rice cultivar 'Nona Bokra'

Many rice cultivars that originated from lower-latitude regions exhibit a strong photoperiod sensitivity (PS) and show extremely late heading under long-day conditions. Under natural day-length conditions during the cropping season in Japan, the indica rice cultivar 'Nona Bokra' from India showed extremely late heading (202 days to heading) compared to the japonica cultivar 'Koshihikari' (105 days), from Japan. To elucidate the genetic factors associated with such extremely late heading, we performed quantitative trait locus (QTL) analyses of heading date using an F(2) population and seven advanced backcross progeny (one BC(1)F(2) and six BC(2)F(2)) derived from a cross between 'Nona Bokra' and 'Koshihikari'. The analyses revealed 12 QTLs on seven chromosomes. The 'Nona Bokra' alleles of all QTLs contributed to an increase in heading date. Digenic interactions were rarely observed between QTLs. Based on the genetic parameters of the QTLs, such as additive effects and percentage of phenotypic variance explained, these 12 QTLs are likely generate a large proportion of the phenotypic variation observed in the heading dates between 'Nona Bokra' and 'Koshihikari'. Comparison of chromosomal locations between heading date QTLs detected in this study and QTLs previously identified in 'Nipponbare' x 'Kasalath' populations revealed that eight of the heading date QTLs were recognized nearby the Hd1, Hd2, Hd3a, Hd4, Hd5, Hd6, Hd9, and Hd13. These results suggest that the strong PS in 'Nona Bokra' was generated mainly by the accumulation of additive effects of particular alleles at previously identified QTLs.

Mapping QTLs influencing rice floral morphology using recombinant inbred lines derived from a cross between Oryza sativa L. and Oryza rufipogon Griff

To understand the genetic basis of floral traits associated with the mating system in rice, we analyzed pistil, stamen and glume traits using a recombinant inbred line population, derived from a cross between an Asian cultivated rice ( Oryza sativa L.), Pei-kuh, and a wild rice ( Oryza rufipogon Griff.), W1944. Quantitative trait loci (QTLs) affecting floral morphology were detected by composite interval mapping using a linkage map constructed using 147 markers, mostly RFLPs. A total of 7, 4, 14 and 6 QTLs were detected for traits related to pistil, stamen, and size and shape of the glume, respectively. Comparison of 31 QTLs affecting these organs revealed ten QTLs affecting the different organs in four adjacent regions on chromosomes 2, 4, 5 and 10, but most QTLs (68%) were located separately on the whole chromosomes. Although four QTLs for stigma breadth, anther length and thickness of lemma and palea explained more than 25% of the total phenotypic variance, most QTLs (87%) had smaller effects. These results suggest that quantitative variation observed for pistil, stamen and glume traits is controlled by several distinct genes with small effects.