Integrating gated recurrent unit in graph neural network to improve infectious disease prediction: an attempt.

Objective: This study focuses on enhancing the precision of epidemic time series data prediction by integrating Gated Recurrent Unit (GRU) into a Graph Neural Network (GNN), forming the GRGNN. The accuracy of the GNN (Graph Neural Network) network with introduced GRU (Gated Recurrent Units) is validated by comparing it with seven commonly used prediction methods. Method: The GRGNN methodology involves multivariate time series prediction using a GNN (Graph Neural Network) network improved by the integration of GRU (Gated Recurrent Units). Additionally, Graphical Fourier Transform (GFT) and Discrete Fourier Transform (DFT) are introduced. GFT captures inter-sequence correlations in the spectral domain, while DFT transforms data from the time domain to the frequency domain, revealing temporal node correlations. Following GFT and DFT, outbreak data are predicted through one-dimensional convolution and gated linear regression in the frequency domain, graph convolution in the spectral domain, and GRU (Gated Recurrent Units) in the time domain. The inverse transformation of GFT and DFT is employed, and final predictions are obtained after passing through a fully connected layer. Evaluation is conducted on three datasets: the COVID-19 datasets of 38 African countries and 42 European countries from worldometers, and the chickenpox dataset of 20 Hungarian regions from Kaggle. Metrics include Average Root Mean Square Error (ARMSE) and Average Mean Absolute Error (AMAE). Result: For African COVID-19 dataset and Hungarian Chickenpox dataset, GRGNN consistently outperforms other methods in ARMSE and AMAE across various prediction step lengths. Optimal results are achieved even at extended prediction steps, highlighting the model's robustness. Conclusion: GRGNN proves effective in predicting epidemic time series data with high accuracy, demonstrating its potential in epidemic surveillance and early warning applications. However, further discussions and studies are warranted to refine its application and judgment methods, emphasizing the ongoing need for exploration and research in this domain.

Arabidopsis mRNA decay landscape shaped by XRN 5'-3' exoribonucleases.

5'-3' exoribonucleases (XRNs) play crucial roles in the control of RNA processing, quality, and quantity in eukaryotes. Although genome-wide profiling of RNA decay fragments is now feasible, how XRNs shape the plant mRNA degradome remains elusive. Here, we profiled and analyzed the RNA degradomes of Arabidopsis wild-type and mutant plants with defects in XRN activity. Deficiency of nuclear XRN3 or cytoplasmic XRN4 activity but not nuclear XRN2 activity greatly altered Arabidopsis mRNA decay profiles. Short excised linear introns and cleaved pre-mRNA fragments downstream of polyadenylation sites were polyadenylated and stabilized in the xrn3 mutant, demonstrating the unique function of XRN3 in the removal of cleavage remnants from pre-mRNA processing. Further analysis of stabilized XRN3 substrates confirmed that pre-mRNA 3' end cleavage frequently occurs after adenosine. The most abundant decay intermediates in wild-type plants include not only the primary substrates of XRN4 but also the products of XRN4-mediated cytoplasmic decay. An increase in decay intermediates with 5' ends upstream of a consensus motif in the xrn4 mutant suggests that there is an endonucleolytic cleavage mechanism targeting the 3' untranslated regions of many Arabidopsis mRNAs. However, analysis of decay fragments in the xrn4 mutant indicated that, except for microRNA-directed slicing, endonucleolytic cleavage events in the coding sequence rarely result in major decay intermediates. Together, these findings reveal the major substrates and products of nuclear and cytoplasmic XRNs along Arabidopsis transcripts and provide a basis for precise interpretation of RNA degradome data.

Nesting the SIRV model with NAR, LSTM and statistical methods to fit and predict COVID-19 epidemic trend in Africa.

OBJECTIVE: Compared with other regions in the world, the transmission characteristics of the COVID-19 epidemic in Africa are more obvious, has a unique transmission mode in this region; At the same time, the data related to the COVID-19 epidemic in Africa is characterized by low data quality and incomplete data coverage, which makes the prediction method of COVID-19 epidemic suitable for other regions unable to achieve good results in Africa. In order to solve the above problems, this paper proposes a prediction method that nests the in-depth learning method in the mechanism model. From the experimental results, it can better solve the above problems and better adapt to the transmission characteristics of the COVID-19 epidemic in African countries. METHODS: Based on the SIRV model, the COVID-19 transmission rate and trend from September 2021 to January 2022 of the top 15 African countries (South Africa, Morocco, Tunisia, Libya, Egypt, Ethiopia, Kenya, Zambia, Algeria, Botswana, Nigeria, Zimbabwe, Mozambique, Uganda, and Ghana) in the accumulative number of COVID-19 confirmed cases was fitted by using the data from Worldometer. Non-autoregressive (NAR), Long-short term memory (LSTM), Autoregressive integrated moving average (ARIMA) models, Gaussian and polynomial functions were used to predict the transmission rate ß in the next 7, 14, and 21 days. Then, the predicted transmission rate ßs were substituted into the SIRV model to predict the number of the COVID-19 active cases. The error analysis was conducted using root-mean-square error (RMSE) and mean absolute percentage error (MAPE). RESULTS: The fitting curves of the 7, 14, and 21 days were consistent with and higher than the original curves of daily active cases (DAC). The MAPE between the fitted and original 7-day DAC was only 1.15% and increased with the longer of predict days. Both the predicted ß and DAC of the next 7, 14, and 21 days by NAR and LSTM nested models were closer to the real ones than other three ones. The minimum RMSEs for the predicted number of COVID-19 active cases in the next 7, 14, and 21 days were 12,974, 14,152, and 12,211 people, respectively when the order of magnitude for was 106, with the minimum MAPE being 1.79%, 1.97%, and 1.64%, respectively. CONCLUSION: Nesting the SIRV model with NAR, LSTM, ARIMA methods etc. through functionalizing ß respectively could obtain more accurate fitting and predicting results than these models/methods alone for the number of confirmed COVID-19 cases in Africa in which nesting with NAR had the highest accuracy for the 14-day and 21-day predictions. The nested model was of high significance for early understanding of the COVID-19 disease burden and preparedness for the response.

Cultivar-specific markers, mutations, and chimerisim of Cavendish banana somaclonal variants resistant to Fusarium oxysporum f. sp. cubense tropical race 4.

BACKGROUND: The selection of tissue culture-derived somaclonal variants of Giant Cavendish banana (Musa spp., Cavendish sub-group AAA) by the Taiwan Banana Research Institute (TBRI) has resulted in several cultivars resistant to Fusarium oxysporum f. sp. cubense tropical race 4 (Foc TR4), a destructive fungus threatening global banana production. However, the mutations in these somaclonal variants have not yet been determined. We performed an RNA-sequencing (RNA-seq) analysis of three TBRI Foc TR4-resistant cultivars: 'Tai-Chiao No. 5' (TC5), 'Tai-Chiao No. 7' (TC7), and 'Formosana' (FM), as well as their susceptible progenitor 'Pei-Chiao' (PC), to investigate the sequence variations among them and develop cultivar-specific markers. RESULTS: A group of single-nucleotide variants (SNVs) specific to one cultivar were identified from the analysis of RNA-seq data and validated using Sanger sequencing from genomic DNA. Several SNVs were further converted into cleaved amplified polymorphic sequence (CAPS) markers or derived CAPS markers that could identify the three Foc TR4-resistant cultivars among 6 local and 5 international Cavendish cultivars. Compared with PC, the three resistant cultivars showed a loss or alteration of heterozygosity in some chromosomal regions, which appears to be a consequence of single-copy chromosomal deletions. Notably, TC7 and FM shared a common deletion region on chromosome 5; however, different TC7 tissues displayed varying degrees of allele ratios in this region, suggesting the presence of chimerism in TC7. CONCLUSIONS: This work demonstrates that reliable SNV markers of tissue culture-derived and propagated banana cultivars with a triploid genome can be developed through RNA-seq data analysis. Moreover, the analysis of sequence heterozygosity can uncover chromosomal deletions and chimerism in banana somaclonal variants. The markers obtained from this study will assist with the identification of TBRI Cavendish somaclonal variants for the quality control of tissue culture propagation, and the protection of breeders' rights.

Identification of MaWRKY40 and MaDLO1 as Effective Marker Genes for Tracking the Salicylic Acid-Mediated Immune Response in Bananas.

Bananas are among the world's most important cash and staple crops but are threatened by various devastating pathogens. The phytohormone salicylic acid (SA) plays a key role in the regulation of plant immune response. Tracking the expression of SA-responsive marker genes under pathogen infection is important in pathogenesis elucidation. However, the common SA-responsive marker genes are not consistently induced in different banana cultivars or different organs. Here, we conducted transcriptome analysis for SA response of a banana cultivar, 'Pei-Chiao' (Cavendish, AAA genome), and identified three genes, MaWRKY40, MaWRKY70, and Downy Mildew Resistant 6 (DMR6)-Like Oxygenase 1 (MaDLO1) that are robustly induced upon SA treatment in both the leaves and roots. Consistent induction of these three genes by SA treatment was also detected in both the leaves and roots of bananas belonging to different genome types such as 'Tai-Chiao No. 7' (Cavendish, AAA genome), 'Pisang Awak' (ABB genome), and 'Lady Finger' (AA genome). Furthermore, the biotrophic pathogen cucumber mosaic virus elicited the expression of MaWRKY40 and MaDLO1 in infected leaves of susceptible cultivars. The hemibiotrophic fungal pathogen Fusarium oxysporum f. sp. cubense tropical race 4 (TR4) also consistently induced the expression of MaWRKY40 and MaDLO1 in the infected roots of the F. oxysporum f. sp. cubense TR4-resistant cultivar. These results indicate that MaWRKY40 and MaDLO1 can be used as reliable SA-responsive marker genes for the study of plant immunity in banana. Revealing SA-responsive marker genes provides a stepping stone for further studies in banana resistance to pathogens.

Widespread Exon Junction Complex Footprints in the RNA Degradome Mark mRNA Degradation before Steady State Translation.

Exon junction complexes (EJCs) are deposited on mRNAs during splicing and displaced by ribosomes during the pioneer round of translation. Nonsense-mediated mRNA decay (NMD) degrades EJC-bound mRNA, but the lack of suitable methodology has prevented the identification of other degradation pathways. Here, we show that the RNA degradomes of Arabidopsis (Arabidopsis thaliana), rice (Oryza sativa), worm (Caenorhabditis elegans), and human (Homo sapiens) cells exhibit an enrichment of 5' monophosphate (5'P) ends of degradation intermediates that map to the canonical EJC region. Inhibition of 5' to 3' exoribonuclease activity and overexpression of an EJC disassembly factor in Arabidopsis reduced the accumulation of these 5'P ends, supporting the notion that they are in vivo EJC footprints. Hundreds of Arabidopsis NMD targets possess evident EJC footprints, validating their degradation during the pioneer round of translation. In addition to premature termination codons, plant microRNAs can also direct the degradation of EJC-bound mRNAs. However, the production of EJC footprints from NMD but not microRNA targets requires the NMD factor SUPPRESSOR WITH MORPHOLOGICAL EFFECT ON GENITALIA PROTEIN7. Together, our results demonstrating in vivo EJC footprinting in Arabidopsis unravel the composition of the RNA degradome and provide a new avenue for studying NMD and other mechanisms targeting EJC-bound mRNAs for degradation before steady state translation.

Engineering Plant Resistance to Tomato Yellow Leaf Curl Thailand Virus Using a Phloem-Specific Promoter Expressing Hairpin RNA.

Transgenic approaches employing RNA interference (RNAi) strategies have been successfully applied to generate desired traits in plants; however, variations between RNAi transgenic siblings and the ability to quickly apply RNAi resistance to diverse cultivars remain challenging. In this study, we assessed the promoter activity of a cauliflower mosaic virus 35S promoter (35S) and a phloem-specific promoter derived from rice tungro bacilliform virus (RTBV) and their efficacy to drive RNAi against the endogenous glutamate-1-semialdehyde aminotransferase gene (GSA) that acts as a RNAi marker, through chlorophyll synthesis inhibition, and against tomato yellow leaf curl Thailand virus (TYLCTHV), a begomovirus (family Geminiviridae) reported to be the prevalent cause of tomato yellow leaf curl disease (TYLCD) in Taiwan. Transgenic Nicotiana benthamiana expressing hairpin RNA of GSA driven by either the 35S or RTBV promoter revealed that RTBV::hpGSA induced stronger silencing along the vein and more uniformed silencing phenotype among its siblings than 35S::hpGSA. Analysis of transgenic N. benthamiana, 35S::hpTYLCTHV, and RTBV::hpTYLCTHV revealed that, although 35S::hpTYLCTHV generated a higher abundance of small RNA than RTBV::hpTYLCTHV, RTBV::hpTYLCTHV transgenic plants conferred better TYLCTHV resistance than 35S::hpTYLCTHV. Grafting of wild-type (WT) scions to TYLCTHV RNAi rootstocks allowed transferable TYLCTHV resistance to the scion. A TYLCTHV-inoculation assay showed that noninfected WT scions were only observed when grafted to RTBV::hpTYLCTHV rootstocks but not 35S::hpTYLCTHV nor WT rootstocks. Together, our findings demonstrate an approach that may be widely applied to efficiently confer TYLCD resistance.

Cucumber mosaic virus-induced gene silencing in banana.

Banana (Musa spp.) is one of the world's most important staple and cash crops. Despite accumulating genetic and transcriptomic data, low transformation efficiency in agronomically important Musa spp. render translational researches in banana difficult by using conventional knockout approaches. To develop tools for translational research in bananas, we developed a virus induced-gene silencing (VIGS) system based on a banana-infecting cucumber mosaic virus (CMV) isolate, CMV 20. CMV 20 genomic RNA 1, 2, and 3, were separately cloned in Agrobacterium pJL89 binary vectors, and a cloning site was introduced on RNA 2 immediately after the 2a open reading frame to insert the gene targeted for silencing. An efficient Agrobacterium inoculation method was developed for banana, which enabled the CMV 20 VIGS vector infection rate to reach 95% in our experiments. CMV 20-based silencing of Musa acuminata cv. Cavendish (AAA group) glutamate 1-semialdehyde aminotransferase (MaGSA) produced a typical chlorotic phenotype and silencing of M. acuminata phytoene desaturase (MaPDS) produced a photobleachnig phenotype. We show this approach efficiently reduced GSA and PDS transcripts to 10% and 18% of the control, respectively. The high infection rate and extended silencing of this VIGS system will provide an invaluable tool to accelerate functional genomic studies in banana.

DCL2- and RDR6-dependent transitive silencing of SMXL4 and SMXL5 in Arabidopsis dcl4 mutants causes defective phloem transport and carbohydrate over-accumulation.

DICER-LIKE (DCL) enzymes process double-stranded RNA into small RNAs that act as regulators of gene expression. Arabidopsis DCL4 and DCL2 each allow the post-transcriptional gene silencing (PTGS) of viruses and transgenes, but primary PTGS-prone DCL4 outcompetes transitive PTGS-prone DCL2 in wild-type plants. This hierarchy likely prevents DCL2 having any detrimental effects on endogenous genes. Indeed, dcl4 mutants exhibit developmental defects and increased sensitivity to genotoxic stress. In this study, the mechanism underlying dcl4 defects was investigated using genetic, biochemical and high-throughput sequencing approaches. We show that the purple phenotype of dcl4 leaves correlates with carbohydrate over-accumulation and defective phloem transport, and depends on the activity of SUPPRESSOR OF GENE SILENCING 3, RNA-DEPENDENT RNA POLYMERASE 6 (RDR6) and DCL2. This phenotype correlates with the downregulation of two genes expressed in the apex and the vasculature, SMAX1-LIKE 4 (SMXL4) and SMXL5, and the accumulation of DCL2- and RDR6-dependent small interfering RNAs derived from these two genes. Supporting a causal effect, smxl4 smxl5 double mutants exhibit leaf pigmentation, enhanced starch accumulation and defective phloem transport, similar to dcl4 plants. Overall, this study elucidates the detrimental action of DCL2 when DCL4 is absent, and indicates that DCL4 outcompeting DCL2 in wild-type plants is crucial to prevent the degradation of endogenous transcripts by DCL2- and RDR6-dependent transitive PTGS.