Sorghum is known as camel among the crops, grown worldwide for food, fodder, and fuel. However, sorghum is highly susceptible to low temperature stress, which greatly affects seed germination, seedling vigor, root architecture, level of cyanogenic glycosides, fertility, and grain yield. Adaptation to low temperature is very crucial for achieving desirable yields under temperate conditions. Here, an association mapping study was conducted using the large global sorghum diversity germplasm accessions to delineate the genetics of early season cold (ESC) and late season frost (LSF) tolerance in sorghum. A total of 10 single nucleotide polymorphisms (SNPs) and 17 quantitative trait loci (QTLs) were identified for ESC and 40 SNPs were identified for LSF. Two ESC tolerance QTLs identified from our study were co-localized with the classical tannin genes Tan1 and Tan2. This study identified probable candidate genes: Sobic.001G157100 (NPH3), Sobic.001G156600 (Lectin receptor-like serine/threonine kinase), and Sobic.006G061100 (SnRK1 gamma sub-unit) as for ESC tolerance and Sobic.006G139900 (UDP-glucoronosyl and UDP-glucosyl transferase), Sobic.002G187400 (Serine/threonine protein kinase), and Sobic.004G333700 (Ca2+/calmodulin-dependent protein phosphatase) for LSF tolerance in sorghum. The identified candidate genes were known to play a major role in seed germinability under cold stress and involved in plant signal transduction and regulation of cold and other biotic and abiotic stresses in crop plants.

Crop pest detection and mitigation remains an extremely challenging task for the farmers. Majority of the pest classification and detection techniques rely on supervised deep learning frameworks that require significant human intervention in labeling the input data, thereby making the down-stream tasks tedious. Therefore, this study presents a self-supervised learning (SSL) approach to classifying 12 types of agricultural insect pests from 9549 RGB images, by leveraging the Bootstrap your own latent (BYOL) algorithm. SSL uses minimal labeling and is indifferent to data augmentations or distortions. Hence, latent representations from pretrained SSL networks could be generalized well for downstream tasks like classification or object detection. For desirable classification of the insect images, the greatest challenges observed were: i) large intra-class variation (the same insect was found with different colors and patterns), and ii) complex background with inconspicuous foreground. Hence, to overcome these issues and aid generalizability of the representations learned through BYOL, entropy-guided segmentation (segments based on texture not color), is proposed as input to the SSL network in this study. Both raw and segmented images were separately fed to two independent BYOL SSL networks, i.e., with ResNet18 and ResNet50 architectures as the backbone. The efficacy of the latent representations for downstream applications was assessed using linear evaluation, and subsequently compared with traditional transfer learning outcomes from ResNet18 and ResNet50. The results indicated that, while ResNet50 backbone intuitively performed better in all cases, SSL aided with entropy-based segmentation led to ~94% classification accuracy compared to raw images (with ~90% maximum accuracy).