Depression recognition using voice-based pre-training model.

The early screening of depression is highly beneficial for patients to obtain better diagnosis and treatment. While the effectiveness of utilizing voice data for depression detection has been demonstrated, the issue of insufficient dataset size remains unresolved. Therefore, we propose an artificial intelligence method to effectively identify depression. The wav2vec 2.0 voice-based pre-training model was used as a feature extractor to automatically extract high-quality voice features from raw audio. Additionally, a small fine-tuning network was used as a classification model to output depression classification results. Subsequently, the proposed model was fine-tuned on the DAIC-WOZ dataset and achieved excellent classification results. Notably, the model demonstrated outstanding performance in binary classification, attaining an accuracy of 0.9649 and an RMSE of 0.1875 on the test set. Similarly, impressive results were obtained in multi-classification, with an accuracy of 0.9481 and an RMSE of 0.3810. The wav2vec 2.0 model was first used for depression recognition and showed strong generalization ability. The method is simple, practical, and applicable, which can assist doctors in the early screening of depression.

[A research on depression recognition based on voice pre-training model].

For the increasing number of patients with depression, this paper proposes an artificial intelligence method to effectively identify depression through voice signals, with the aim of improving the efficiency of diagnosis and treatment. Firstly, a pre-training model called wav2vec 2.0 is fine-tuned to encode and contextualize the speech, thereby obtaining high-quality voice features. This model is applied to the publicly available dataset - the distress analysis interview corpus-wizard of OZ (DAIC-WOZ). The results demonstrate a precision rate of 93.96%, a recall rate of 94.87%, and an F1 score of 94.41% for the binary classification task of depression recognition, resulting in an overall classification accuracy of 96.48%. For the four-class classification task evaluating the severity of depression, the precision rates are all above 92.59%, the recall rates are all above 92.89%, the F1 scores are all above 93.12%, and the overall classification accuracy is 94.80%. The research findings indicate that the proposed method effectively enhances classification accuracy in scenarios with limited data, exhibiting strong performance in depression identification and severity evaluation. In the future, this method has the potential to serve as a valuable supportive tool for depression diagnosis.

Challenging Isodimorphism Concepts: Formation of Three Crystalline Phases in Poly(hexamethylene-ran-octamethylene carbonate) Copolymers.

In this work, poly(hexamethylene-ran-octamethylene carbonate) copolycarbonates were synthesized by melt polycondensation in a wide range of compositions. The copolymers displayed some of the characteristic isodimorphic thermal behavior, such as crystallization for all the compositions and a pseudoeutectic behavior of the melting temperature (Tm) versus composition. The pseudoeutectic point was located at 33 mol % poly(octamethylene carbonate) (POC) content (i.e., corresponding to the PH67O33C copolymer). Surprisingly, the crystallinities (Xc) for a wide range of copolymer compositions were higher than those of the parent components, a phenomenon that has not been observed before in isodimorphic random copolymers. The structural characterization, performed by wide-angle X-ray scattering (WAXS) and small-angle X-ray scattering experiments, revealed unexpected results depending on composition. On the one hand, the poly(hexamethylene carbonate) (PHC)- and POC-rich copolymers crystallize in PHC- and POC-type crystals, as expected. Moreover, upon cooling and heating, in situ WAXS experiments evidenced that these materials undergo reversible solid-solid transitions [δ-α (PHC) and δ-α-ß (POC)] present in the parent components but at lower temperatures. On the other hand, a novel behavior was found for copolymers with 33-73 mol % POC (including the pseudoeutectic point), which are those with higher crystallinities than the parent components. For these copolymers, a new crystalline phase that is different from that of both homopolymers was observed. The in situ WAXS results for these copolymers confirmed that this novel phase is stable upon cooling and heating and does not show any crystallographic feature of the parent components or their solid-solid transitions. FTIR experiments confirmed this behavior, revealing that the new phase adopts a polyethylene-like chain conformation that differs from the trans-dominant ones exhibited by the parent components. This finding challenges the established concepts of isodimorphism and questions whether a combination of crystallization modes (isodimorphism and isomorphism) is possible in the same family of random copolymers just by changing the composition.

Bulk segregant analysis coupled with transcriptomics and metabolomics revealed key regulators of bacterial leaf blight resistance in rice.

BACKGROUND: Bacterial leaf blight (BLB) is a highly destructive disease, causing significant yield losses in rice (Oryza sativa). Genetic variation is contemplated as the most effective measure for inducing resistance in plants. The mutant line T1247 derived from R3550 (BLB susceptible) was highly resistant to BLB. Therefore, by utilizing this valuable source, we employed bulk segregant analysis (BSA) and transcriptome profiling to identify the genetic basis of BLB resistance in T1247. RESULTS: The differential subtraction method in BSA identified a quantitative trait locus (QTL) on chromosome 11 spanning a 27-27.45 Mb region with 33 genes and 4 differentially expressed genes (DEGs). Four DEGs (P < 0.01) with three putative candidate genes, OsR498G1120557200, OsR498G1120555700, and OsR498G1120563600,0.01 in the QTL region were identified with specific regulation as a response to BLB inoculation. Moreover, transcriptome profiling identified 37 resistance analogs genes displaying differential regulation. CONCLUSIONS: Our study provides a substantial addition to the available information regarding QTLs associated with BLB, and further functional verification of identified candidate genes can broaden the scope of understanding the BLB resistance mechanism in rice.

Transcriptomic and methylomic analyses provide insights into the molecular mechanism and prediction of heterosis in rice.

Heterosis has been widely used in multiple crops. However, the molecular mechanism and prediction of heterosis remains elusive. We generated five F1 hybrids [four showing better-parent heterosis (BPH) and one showing mid-parent heterosis], and performed the transcriptomic and methylomic analyses to identify the candidate genes for BPH and explore the molecular mechanism of heterosis and the potential predictors for heterosis. Transcriptomic results showed that most of the differentially expressed genes shared in the four better-parent hybrids were significantly enriched into the terms of molecular function, and the additive and dominant effects played crucial roles for BPH. DNA methylation level, especially in CG context, significantly and positively correlated with grain yield per plant. The ratios of differentially methylated regions in CG context in exons to transcription start sites between the parents exhibited significantly negative correlation with the heterosis levels of their hybrids, as was further confirmed in 24 pairwise comparisons of other rice lines, implying that this ratio could be a feasible predictor for heterosis level, and this ratio of less than 5 between parents in early growth stages might be a critical index for judging that their F1 hybrids would show BPH. Additionally, we identified some important genes showing differential expression and methylation, such as OsDCL2, Pi5, DTH2, DTH8, Hd1 and GLW7 in the four better-parent hybrids as the candidate genes for BPH. Our findings helped shed more light on the molecular mechanism and heterosis prediction.

Evaluation of DNA adduct damage using G-quadruplex-based DNAzyme.

Toxicity assessment is a major problem in pharmaceutical candidates and industry chemicals development. However, due to the lack of practical analytical methods for DNA adduct analysis, the safety evaluation of drug and industry chemicals was severely limited. Here, we develop a DNAzyme-based method to detect DNA adduct damage for toxicity assessment of drugs and chemicals. Among 18 structural variants of G4 DNAzyme, EA2 DNAzyme exhibits an obvious DNA damaging effect of styrene oxide (SO) due to its unstable structure. The covalent binding of SO to DNAzyme disrupts the Hoogsteen hydrogen bonding sites of G-plane guanines and affects the formation of the G4 quadruplex. DNA damage chemicals reduce the peroxidase activity of the G4 DNAzyme to monitor the DNA adduct damage by disrupting the structural integrity of the G4 DNAzyme. Our method for genotoxic assessment of pharmaceutical candidates and industrial chemicals can elucidate the complex chemical pathways leading to toxicity, predict toxic effects of chemicals, and evaluate possible risks to human health.

The grain yield regulator NOG1 plays a dual role in latitudinal adaptation and cold tolerance during rice domestication.

Rice originated in tropical and subtropical regions and is distributed worldwide. Low temperature is one of the most critical abiotic stresses affecting grain yield and geographical distribution of rice. It is vital to elucidate the molecular mechanism of chilling tolerance in rice for ensuring cereals production. Previously we isolated the domestication-related gene NOG1 which affects rice grain number and yield. In this study, we specified that rice varieties harboring high-yielding NOG1 allele are more distributed in low-latitude regions. Additionally, we observed NOG1 influences the chilling tolerance of rice. Through genome-wide transcriptional analysis after cold treatment at 10°C, there were 717 differentially expressed genes (DEGs) in nog1 near-isogenic lines compared with the control Guichao 2, including 432 up-regulated DEGs and 284 down-regulated DEGs. Gene ontology annotations and KEGG enrichment analysis of DEGs showed that various biological processes and signaling pathways were related to cold stress, such as lipid metabolism and genetic information processing. These results provide new insights into the mechanism of chilling tolerance in rice and the molecular basis of environmental adaptation during rice domestication.

Simple, rapid, and visual electrochemiluminescence sensor for on-site catechol analysis.

Environmental pollution caused by aromatic compounds such as catechol (Cat) has become a major issue for human health. However, there is no simple, rapid, and low-cost method for on-site monitoring of Cat. Here, based on ECL quenching mechanism, we develop a simple, rapid and visual mesoporous silica (MSNs)-electrochemiluminescence (ECL) sensor for on-site monitoring of Cat. The mechanism of ECL quenching is due to the interaction between Cat and Ru(bpy)3 2+* and the interactions between the oxidation products of Cat and DBAE. MSNs films with ordered perpendicular mesopore channels exhibit an amplification effect of ECL intensity due to the negatively charged pore channel. There is a good linear relationship between ECL intensity and Cat concentration in the range of 10 ∼ 1000 µM with the limit of detection (LOD) of 9.518 µM (R 2 = 0.99). The on-site sensor is promising to offer new opportunities for pharmaceuticals analysis, on-site monitoring, and exposure risk assessment.

Identification of Heterotic Groups and Patterns Based on Genotypic and Phenotypic Characteristics Among Rice Accessions of Diverse Origins.

Identification of the right parental combinations to maximize heterosis is the major goal of hybrid breeding, which could be achieved through identification of heterotic groups. The main objective of this study was to identify promising heterotic groups for future rice breeding programs. A collection of 359 rice genotypes of diverse origins of China and abroad, composed of inbreds, maintainers, restorers, and temperature-sensitive genic male sterile (TGMS) lines were genotyped using 10K SNP chips. The SNP data set was subjected to genomic analyses for estimation of genetic divergence and diversity. Significant variations were observed in the germplasm with the identification of six different genetic groups. These lines were assigned to the genetic groups independent of their origin. Taking an account of commercially used heterotic groups present in each cluster, three cytoplasmic male sterile (CMS) lines and 14 inbred and restorer lines with moderate to high genetic distances selected from five heterotic patterns were crossed and obtained 42 F1 hybrids. A total of 14 hybrids were found with significant maximum mid- and better-parent heterosis, namely, TaifengA × Guang122, TaifengA × Wushansimiao, and TaifengA × Minghui63 for earliness; Guang8A × Huazhan for dwarf stature; and Guang8A × Huanghuzhan-1, TaifengA × Yuexiangzhan, Guang8A × Minhui3301, TianfengA × Guang122, Guang8A × Yahui2115, TianfengA × Huanghuazhan, TianfengA × Minghui63, TianfengA × Minhui3301, TaifengA × Gui99, and Guang8A × Yuenongsimiao for yield and yield-related traits. Mid-parent and better-parent heterotic F1 hybrids were in positive correlation with the genetic distances as that manifested by commercially used heterotic groups, encouraging the use of genotypic data for identification of heterotic groups. Our study provides an informative strategy for the development of early maturing, lodging resistant and high-yielding commercial hybrids and cultivars in future heterosis breeding programs.

QTL mapping using an ultra-high-density SNP map reveals a major locus for grain yield in an elite rice restorer R998.

To dissect the genetic basis of yield formation in restorer line of hybrid rice, we conducted QTL analysis for 6 yield traits including panicles per plant (PPP), grains per panicle (GPP), grain yield per plant (GY), thousand-grain weight (TGW), above-ground biomass (AGB), and harvest index (HI) using SNP markers in a recombinant inbred lines (RILs) population derived from a cross between a tropical japonica inbred Francis and an elite indica restorer Guanghui 998 (R998). A total of 26 QTLs were detected using a high density genetic map consisting of 3016 bin markers. Nineteen out of the 26 QTL alleles from R998 had a beneficial effect on yield traits. Most of the QTLs were co-located with previously reported rice QTLs. qAGB6 and qHI9, controlling AGB and HI respectively, were detected as novel QTLs. Four QTLs for GY were repeatedly detected across two years, with all the beneficial alleles from R998. Notably, qGY8 explained over 20% of the yield variance in both years. Moreover, qGY8 together with qTGW8 and qHI8 formed a QTL cluster. Markers tightly linked with qGY8 were developed. Cloning of qGY8 will facilitate its further exploitation in high-yield breeding.

Transcriptomic Analysis Reveals New Insights into High-Temperature-Dependent Glume-Unclosing in an Elite Rice Male Sterile Line.

Glume-unclosing after anthesis is a widespread phenomenon in hybrid rice and also a maternal hereditary trait. The character of Glume-unclosing in rice male sterile lines also seriously influences germination rate and the commercial quality of hybrid rice seeds. We validated that the type of glume-unclosing after anthesis in the elite rice thermo-sensitive genic male sterile (TGMS) line RGD-7S was caused by high temperature. Transcriptomic sequencing of rice panicles was performed to explore the change of transcript profiles under four conditions: pre- and post-anthesis under high temperature (HRGD0 and HRGD1), and pre- and post-anthesis under low temperature (LRGD0 and LRGD1). We identified a total of 14,540 differentially expressed genes (DEGs) including some heat shock factors (HSFs) across the four samples. We found that more genes were up-regulated than down-regulated in the sample pair HRGD1vsHRGD0. These up-regulated genes were significantly enriched in the three biological processes of carbohydrate metabolism, response to water and cell wall macromolecular metabolism. Simultaneously, we also found that the HSF gene OsHsfB1 was specially up-regulated in HRGD1vsHRGD0. However, the down-regulated DEGs in LRGD1vsLRGD0 were remarkably clustered in the biological process of carbohydrate metabolism. This suggests that carbohydrate metabolism may play a key role in regulation of glume-unclosing under high temperature in RGD-7S. We also analyzed the expression pattern of genes enriched in carbohydrate metabolism and several HSF genes under different conditions and provide new insights into the cause of rice glume-unclosing.

Genome-wide DNA polymorphism in the indica rice varieties RGD-7S and Taifeng B as revealed by whole genome re-sequencing.

Next-generation sequencing technologies provide opportunities to further understand genetic variation, even within closely related cultivars. We performed whole genome resequencing of two elite indica rice varieties, RGD-7S and Taifeng B, whose F1 progeny showed hybrid weakness and hybrid vigor when grown in the early- and late-cropping seasons, respectively. Approximately 150 million 100-bp pair-end reads were generated, which covered ∼86% of the rice (Oryza sativa L. japonica 'Nipponbare') reference genome. A total of 2,758,740 polymorphic sites including 2,408,845 SNPs and 349,895 InDels were detected in RGD-7S and Taifeng B, respectively. Applying stringent parameters, we identified 961,791 SNPs and 46,640 InDels between RGD-7S and Taifeng B (RGD-7S/Taifeng B). The density of DNA polymorphisms was 256.8 SNPs and 12.5 InDels per 100 kb for RGD-7S/Taifeng B. Copy number variations (CNVs) were also investigated. In RGD-7S, 1989 of 2727 CNVs were overlapped in 218 genes, and 1231 of 2010 CNVs were annotated in 175 genes in Taifeng B. In addition, we verified a subset of InDels in the interval of hybrid weakness genes, Hw3 and Hw4, and obtained some polymorphic InDel markers, which will provide a sound foundation for cloning hybrid weakness genes. Analysis of genomic variations will also contribute to understanding the genetic basis of hybrid weakness and heterosis.

Genetic and cytological analysis of a novel type of low temperature-dependent intrasubspecific hybrid weakness in rice.

Hybrid weakness (HW) is an important postzygotic isolation which occurs in both intra- and inter-specific crosses. In this study, we described a novel low temperature-dependent intrasubspecific hybrid weakness in the F1 plants derived from the cross between two indica rice varieties Taifeng A and V1134. HW plants showed growth retardation, reduced panicle number and pale green leaves with chlorotic spots. Cytological assay showed that there were reduced cell numbers, larger intercellular spaces, thicker cell walls, and abnormal development of chloroplast and mitochondria in the mature leaves from HW F1 plants in comparison with that from both of the parental lines. Genetic analysis revealed that HW was controlled by two complementary dominant genes Hw3 from V1134 and Hw4 from Taifeng A. Hw3 was mapped in a 136 kb interval between the markers Indel1118 and Indel1117 on chromosome 11, and Hw4 was mapped in the region of about 15 cM between RM182 and RM505 on chromosome 7, respectively. RT-PCR analysis revealed that only LOC_Os11g44310, encoding a putative calmodulin-binding protein (OsCaMBP), differentially expressed among Taifeng A, V1134 and their HW F1. No recombinant was detected using the markers designed based on the sequence of LOC_Os11g44310 in the BC1F2 (Taifeng A//Taifeng A/V1134) population. Hence, LOC_Os11g44310 was probably the candidate gene of Hw3. Gene amplification suggested that LOC_Os11g44310 was present in V1134 and absent in Taifeng A. BLAST search revealed that LOC_Os11g44310 had one copy in the japonica genomic sequence of Nipponbare, and no homologous sequence in the indica reference sequence of 9311. Our results indicate that Hw3 is a novel gene for inducing hybrid weakness in rice.

Transgene flow to hybrid rice and its male-sterile lines.

Gene flow from genetically modified (GM) crops to the same species or wild relatives is a major concern in risk assessment. Transgenic rice with insect and/or disease resistance, herbicide, salt and/or drought tolerance and improved quality has been successfully developed. However, data on rice gene flow from environmental risk assessment studies are currently insufficient for the large-scale commercialization of GM rice. We have provided data on the gene flow frequency at 17 distances between a GM japonica line containing the bar gene as a pollen donor and two indica hybrid rice varieties and four male-sterile (ms) lines. The GM line was planted in a 640 m2 in an isolated experimental plot (2.4 ha), which simulates actual conditions of rice production with pollen competition. Results showed that: (1) under parallel plantation at the 0-m zone, the transgene flow frequency to the ms lines ranged from 3.145 to 36.116% and was significantly higher than that to hybrid rice cultivars (0.037-0.045%). (2) Gene flow frequency decreased as the distance increased, with a sharp cutoff point at about 1-2 m; (3) The maximum distance of transgene flow was 30-40 m to rice cultivars and 40-150 m to ms lines. We believe that these data will be useful for the risk assessment and management of transgenic rice lines, especially in Asia where 90% of world's rice is produced and hybrid rice varieties are extensively used.

A large-scale field study of transgene flow from cultivated rice (Oryza sativa) to common wild rice (O. rufipogon) and barnyard grass (Echinochloa crusgalli).

The introgression of transgenes into wild relatives or weeds through pollen-mediated gene flow is a major concern in environmental risk assessment of transgenic crops. A large-scale (1.3-1.8 ha) rice gene flow study was conducted using transgenic rice containing the bar gene as a pollen donor and Oryza rufipogon as a recipient. There was a high frequency of transgene flow (11%-18%) at 0-1 m, with a steep decline with increasing distance to a detection limit of 0.01% by 250 m. To our knowledge, this is the highest frequency and longest distance of gene flow from transgenic rice to O. rufipogon reported so far. On the basis of these data, an adequate isolation distance from both conventional and transgenic rice should be taken for in situ conservation of common wild rice. Meanwhile, there is no evidence of transgene introgression into barnyard grass, even when it has coexisted with transgenic rice containing the bar gene for five successive years. Thus, the environmental risk of gene flow to this weedy species is of little concern.