A Shallow Autoencoder Framework for Epileptic Seizure Detection in EEG Signals.

This paper presents a trainable hybrid approach involving a shallow autoencoder (AE) and a conventional classifier for epileptic seizure detection. The signal segments of a channel of electroencephalogram (EEG) (EEG epochs) are classified as epileptic and non-epileptic by employing its encoded AE representation as a feature vector. Analysis on a single channel-basis and the low computational complexity of the algorithm allow its use in body sensor networks and wearable devices using one or few EEG channels for wearing comfort. This enables the extended diagnosis and monitoring of epileptic patients at home. The encoded representation of EEG signal segments is obtained based on training the shallow AE to minimize the signal reconstruction error. Extensive experimentation with classifiers has led us to propose two versions of our hybrid method: (a) one yielding the best classification performance compared to the reported methods using the k-nearest neighbor (kNN) classifier and (b) the second with a hardware-friendly architecture and yet with the best classification performance compared to other reported methods in this category using a support-vector machine (SVM) classifier. The algorithm is evaluated on the Children's Hospital Boston, Massachusetts Institute of Technology (CHB-MIT), and University of Bonn EEG datasets. The proposed method achieves 98.85% accuracy, 99.29% sensitivity, and 98.86% specificity on the CHB-MIT dataset using the kNN classifier. The best figures using the SVM classifier for accuracy, sensitivity, and specificity are 99.19%, 96.10%, and 99.19%, respectively. Our experiments establish the superiority of using an AE approach with a shallow architecture to generate a low-dimensionality yet effective EEG signal representation capable of high-performance abnormal seizure activity detection at a single-channel EEG level and with a fine granularity of 1 s EEG epochs.

Comprehensive analysis of potato (Solanum tuberosum) PYL genes highlights their role in stress responses.

Abscisic acid (ABA) regulates plant development, seed germination, and stress responses. The PYR1-like (PYL) proteins are essential for ABA signalling. However, the evolution and expression of PYL genes in potato (Solanum tuberosum ) remain poorly understood. Here, we analysed and identified 17 PYL genes in the potato genome, which were categorised into three groups based on phylogenetic analysis. These genes are distributed across nine chromosomes with predicted proteins subcellar localisation primarily in the cytoplasm. These StPYLs revealed conserved exon structures and domains among the groups. Promoter region analysis indicated hormone and stress-related elements in all StPYL s. Protein-protein interactions and microRNA networks predicted that the interactions of StPYLs are crucial components of ABA signalling, underlining their pivotal role in stress management and growth regulation in potato. Expression profiling across different tissues and under various stresses revealed their varied expression pattern. Further, we validated the expression pattern of selected StPYLs through reverse transcription quantitative PCR under drought, salt, and Phytophthora infestans stresses. This revealed consistent upregulation of StPYL6 in these stresses, while StPYL11 exhibited significant downregulation over time. Other genes showed downregulation under drought and salt stresses while upregulation under P. infestans . Overall, our results suggested the potential role of PYL genes in abiotic and biotic stresses.

Comparative global profiling of Perilla leaf and stem via transcriptomics and metabolomics.

Perilla (Perilla frutescens L.) is a time-honored herbal plant with widespread applications in both medicine and culinary practices around the world. Profiling the essential organs and tissues with medicinal significance on a global scale offers valuable insights for enhancing the yield of desirable compounds in Perilla and other medicinal plants. In the present study, genome-wide RNA-sequencing (RNA-seq) and assessing the global spectrum of metabolites were carried out in the two major organs/tissues of stem (PfST) and leaf (PfLE) in Perilla. The results showed a total of 18,490 transcripts as the DEGs (differentially expressed genes) and 144 metabolites as the DAMs (differentially accumulated metabolites) through the comparative profiling of PfST vs PfLE, and all the DEGs and DAMs exhibited tissue-specific trends. An association analysis between the transcriptomics and metabolomics revealed 14 significantly enriched pathways for both DEGs and DAMs, among which the pathways of Glycine, serine and threonine metabolism (ko00260), Glyoxylate and dicarboxylate metabolism (ko00630), and Glucagon signaling pathway (ko04922) involved relatively more DEGs and DAMs. The results of qRT-PCR assays of 18 selected DEGs confirmed the distinct tissue-specific characteristics of all identified DEGs between PfST and PfLE. Notably, all eight genes associated with the flavonoid biosynthesis/metabolism pathways exhibited significantly elevated expression levels in PfLE compared to PfST. This observation suggests a heightened accumulation of metabolites related to flavonoids in Perilla leaves. The findings of this study offer a comprehensive overview of the organs and tissues in Perilla that have medicinal significance.

Exploring regulatory network of icariin synthesis in Herba Epimedii through integrated omics analysis.

Herba Epimedii's leaves are highly valued in traditional Chinese medicine for their substantial concentration of flavonoids, which play a crucial role in manifesting the plant's therapeutic properties. This study investigated the metabolomic, transcriptomic and proteomic profiles of leaves from two Herba Epimedii cultivars, Epipremnum sagittatum (J) and Epipremnum pubescens (R), at three different developmental stages. Metabolite identification and analysis revealed a total of 1,412 and 1,421 metabolites with known structures were found. Flavonoids made up of 33%, including 10 significant accumulated icariin analogues. Transcriptomic analysis unveiled totally 41,644 differentially expressed genes (DEGs) containing five encoded genes participated in icariin biosynthesis pathways. Totally, 9,745 differentially expressed proteins (DEPs) were found, including Cluster-47248.2.p1 (UDP-glucuronosy/UDP-glucosyltransferase), Cluster-30441.2.p1 (O-glucosyltransferase), and Cluster-28344.9.p1 (anthocyanidin 3-O-glucoside 2 "-O-glucosyltransferase-like) through proteomics analysis which are involved to icariin biosynthesis. Protein-protein interaction (PPI) assay exhibited, totally 12 proteins showing a strong relationship of false discovery rate (FDR) <0.05 with these three proteins containing 2 leucine-rich repeat receptor kinase-like protein SRF7, and 5 methyl jasmonate esterase 1. Multi-omics connection networks uncovered 237 DEGs and 72 DEPs exhibited significant associations with the 10 icariin analogues. Overall, our integrated omics approach provides comprehensive insights into the regulatory network underlying icariin synthesis in Herba Epimedii, offering valuable resources for further research and development in medicinal plant cultivation and pharmaceutical applications.

Classifying Single Channel Epileptic EEG data based on Sparse Representation using Shallow Autoencoder.

Patient independent epileptic seizure detection algorithm for scalp electroencephalogram (EEG) data is pro- posed in this paper. Principal motivation of this work is to integrate neural and conventional machine learning methods to develop a classification system which can advance the current wearable health systems in terms of computational complexity and accuracy. Being based on processing a single channel EEG processing, the approach is suitable for usage with small wireless sensors. A shallow autoencoder model is utilized for sparse representation of the EEG signal followed by k-nearest neighbor (kNN) classifier to categorize the data as epileptic or non-epileptic. Using a single EEG channel an optimum sparsity level is explored in the encoded sample. Attaining an accuracy, sensitivity and specificity of 98.85%, 99.29% and 98.86% respectively, for CHB-MIT scalp EEG database, proposed classification method outperforms state of- the-art seizure detection methodologies. Experiments has shown that this performance was possible by using a sparsity level of 4 in the auto-encoder. Furthermore, use of shallow learning instead of deep learning approach for generation of sparse but effective representation is computationally lighter than many other feature extraction and preprocessing methods.