SY-Net: A Rice Seed Instance Segmentation Method Based on a Six-Layer Feature Fusion Network and a Parallel Prediction Head Structure.

During the rice quality testing process, the precise segmentation and extraction of grain pixels is a key technique for accurately determining the quality of each seed. Due to the similar physical characteristics, small particles and dense distributions of rice seeds, properly analysing rice is a difficult problem in the field of target segmentation. In this paper, a network called SY-net, which consists of a feature extractor module, a feature pyramid fusion module, a prediction head module and a prototype mask generation module, is proposed for rice seed instance segmentation. In the feature extraction module, a transformer backbone is used to improve the ability of the network to learn rice seed features; in the pyramid fusion module and the prediction head module, a six-layer feature fusion network and a parallel prediction head structure are employed to enhance the utilization of feature information; and in the prototype mask generation module, a large feature map is used to generate high-quality masks. Training and testing were performed on two public datasets and one private rice seed dataset. The results showed that SY-net achieved a mean average precision (mAP) of 90.71% for the private rice seed dataset and an average precision (AP) of 16.5% with small targets in COCO2017. The network improved the efficiency of rice seed segmentation and showed excellent application prospects in performing rice seed quality testing.

Rice Seed Purity Identification Technology Using Hyperspectral Image with LASSO Logistic Regression Model.

Hyperspectral technology is used to obtain spectral and spatial information of samples simultaneously and demonstrates significant potential for use in seed purity identification. However, it has certain limitations, such as high acquisition cost and massive redundant information. This study integrates the advantages of the sparse feature of the least absolute shrinkage and selection operator (LASSO) algorithm and the classification feature of the logistic regression model (LRM). We propose a hyperspectral rice seed purity identification method based on the LASSO logistic regression model (LLRM). The feasibility of using LLRM for the selection of feature wavelength bands and seed purity identification are discussed using four types of rice seeds as research objects. The results of 13 different adulteration cases revealed that the value of the regularisation parameter was different in each case. The recognition accuracy of LLRM and average recognition accuracy were 91.67-100% and 98.47%, respectively. Furthermore, the recognition accuracy of full-band LRM was 71.60-100%. However, the average recognition accuracy was merely 89.63%. These results indicate that LLRM can select the feature wavelength bands stably and improve the recognition accuracy of rice seeds, demonstrating the feasibility of developing a hyperspectral technology with LLRM for seed purity identification.

Computational and functional characterization of four SNPs in the SOST locus associated with osteoporosis.

The SOST gene encodes sclerostin, a C-terminal cysteine knot-like domain containing key negative regulator of osteoblastic bone formation that inhibits LRP5/6-mediated canonical Wnt signaling. Numerous single nucleotide polymorphisms (SNPs) in the SOST locus are firmly associated with bone mineral density (BMD) and fracture in genome-wide association studies (GWAS) and candidate gene association studies. However, the validation and mechanistic elucidation of causal genetic variants, especially for SNPs located beyond the promoter-proximal region, remain largely unresolved. By employing computational and experimental approaches, here we identify four SNPs rs1230399, rs7220711, rs1107748 and rs75901553 as functional variants which display allelic variation in SOST gene expression. The osteoporosis associated SNP rs1230399 in the SOST distal upstream regulatory region shows FOXA1 binding activity with subsequent transinactivation in a T allele-specific manner. The BMD GWAS lead SNPs rs7220711 and rs1107748 both reside in the 52-kb regulatory element deletion 35-kb downstream of the SOST gene which leads to Van Buchem disease. The rs7220711-A has a higher affinity for the transcriptional repressors MAFF or MAFK homodimers than rs7220711-G, while rs1107748 confers C allele specific transcriptional enhancer activity via a CTCF binding element. The variant rs75901553 C>T located in a conserved site of the SOST 3' UTR abolishes a target binding site for miR-98-5p which is negatively responsive to parathyroid hormone or 17ß-estradiol in osteoblastic cell lines. Our findings uncover the biological consequences of four independent genetic variants in the SOST region and their important roles in SOST expression via diverse mechanisms, providing new insights into the genetics and molecular pathogenesis of osteoporosis.

Computational Characterization of Osteoporosis Associated SNPs and Genes Identified by Genome-Wide Association Studies.

OBJECTIVES: Genome-wide association studies (GWASs) have revealed many SNPs and genes associated with osteoporosis. However, influence of these SNPs and genes on the predisposition to osteoporosis is not fully understood. We aimed to identify osteoporosis GWASs-associated SNPs potentially influencing the binding affinity of transcription factors and miRNAs, and reveal enrichment signaling pathway and "hub" genes of osteoporosis GWAS-associated genes. METHODS: We conducted multiple computational analyses to explore function and mechanisms of osteoporosis GWAS-associated SNPs and genes, including SNP conservation analysis and functional annotation (influence of SNPs on transcription factors and miRNA binding), gene ontology analysis, pathway analysis and protein-protein interaction analysis. RESULTS: Our results suggested that a number of SNPs potentially influence the binding affinity of transcription factors (NFATC2, MEF2C, SOX9, RUNX2, ESR2, FOXA1 and STAT3) and miRNAs. Osteoporosis GWASs-associated genes showed enrichment of Wnt signaling pathway, basal cell carcinoma and Hedgehog signaling pathway. Highly interconnected "hub" genes revealed by interaction network analysis are RUNX2, SP7, TNFRSF11B, LRP5, DKK1, ESR1 and SOST. CONCLUSIONS: Our results provided the targets for further experimental assessment and further insight on osteoporosis pathophysiology.