Subcellular Localization Prediction Research Articles

Subtractive genome mining in Xanthomonas citri pv. citri strain 306 for identifying novel drug target proteins coupled with in-depth protein-protein interaction and coevolution analysis - A leap towards prospective drug design.

Citrus canker poses a serious threat to a highly significant citrus fruit crop, this disease caused by one of the most destructive bacterial plant pathogens Xanthomonas citri pv. citri (Xcc). Bacterial plant diseases significantly reduce crop yields worldwide, making it more difficult to supply the growing food demand. The high levels of antibiotic resistance in Xcc strains are diminishing the efficacy of current control measures, necessitating the exploration of novel therapeutic targets to address the escalating antimicrobial resistance trend. Genome subtraction approach along with protein-protein network and coevolution analysis were used to identify potential drug targets in Xcc stain 306. The study involved retrieving the Xcc proteome from the UniProt database, eliminating paralogous proteins using CD-HIT (80% identity cutoff), and selecting nonhomologous proteins through BLASTp (e-value <0.005). Essential proteins were identified using BLAST against the DEG (e-value cutoff 0.00001). 750 essential proteins were identified that are nonhomologous to citrus plant. Subsequent analyses included metabolic pathway assessment, subcellular localization prediction, and druggability analysis. Protein network analysis, coevolution analysis, protein active site identification was also performed. In conclusion, this study identified eight potential drug targets (GlmU, CheA, RmlD, GspE, FleQ, RpoN, Shk, SecB), highlighting RpoN, FleQ, and SecB as unprecedented targets for Xcc. These findings may contribute to the development of novel antimicrobial agents in the future that can efficiently control citrus canker disease.

Genome-wide characterization and expression pattern analysis of Xanthoceras sorbifolium FAD at different growth stages

ABSTRACT Xanthoceras sorbifolium is a crucial oil tree species in northern China. Additionally, fatty acid constituents in its fruits possess remarkable medicinal value. Hence, probing the regulatory role of FAD family members in fatty acid components accumulation in X. sorbifolium fruits is crucial. Drawing on X. sorbifolium genome data, 22 XsFAD members distributed across 10 chromosomes were identified. The physicochemical properties and subcellular localization predictions thereof were analyzed. Phylogenetic and conserved domain analyses demonstrated that same subfamily members are highly conserved in both gene structure and conserved domains. Cis-acting element analysis indicated that XsFAD members are highly susceptible to light. Furthermore, transcriptome data and qRT-PCR findings corroborated that XsFAD members play an active regulatory role in X. sorbifolium fruit ripening and fatty acid metabolism. The bioinformatic analysis and preliminary verification of regulatory functions of XsFAD members establish conditions for in-depth research on the functions of key XsFAD in the future. The study findings will be conducive to determine the accumulation of fatty acids substances in X. sorbifolium and screening of germplasm resources with superior traits.

BiGM-lncLoc: Bi-level Multi-Graph Meta-Learning for Predicting Cell-Specific Long Noncoding RNAs Subcellular Localization.

The precise spatiotemporal expression of long noncoding RNAs (lncRNAs) plays a pivotal role in biological regulation, and aberrant expression of lncRNAs in different subcellular localizations has been intricately linked to the onset and progression of a variety of cancers. Computational methods provide effective means for predicting lncRNA subcellular localization, but current studies either ignore cell line and tissue specificity or the correlation and shared information among cell lines. In this study, we propose a novel approach, BiGM-lncLoc, treating the prediction of lncRNA subcellular localization across cell lines as a multi-graph meta-learning task. Our investigation involves two categories of data: the localization data of nucleotide sequences in different cell lines and cell line expression data. BiGM-lncLoc comprises a cell line-specific optimization network learning specific knowledge from cell line expression data and a graph neural network optimized across cell lines. Subsequently, the specific and shared knowledge acquired through bi-level optimization is applied to a new cell-line prediction task without the need for re-training or fine-tuning. Additionally, through key feature analysis of the impact of different nucleotide combinations on the model, we confirm the necessity of cell line-specific studies based on correlation analysis. Finally, experiments conducted on various cell lines with different data sizes indicate that BiGM-lncLoc outperforms other methods in terms of prediction accuracy, with an average accuracy of 97.7%. After removing overlapping samples to ensure data independence for each cell line, the accuracy ranged from 82.4% to 94.7%, still surpassing existing models. Our code can be found at https://github.com/BioCL1/BiGM-lncLoc .

A Comprehensive Analysis In Silico of KCS Genes in Maize Revealed Their Potential Role in Response to Abiotic Stress.

β-ketoacyl-CoA synthase (KCS) enzymes play a pivotal role in plants by catalyzing the first step of very long-chain fatty acid (VLCFA) biosynthesis. This process is crucial for plant development and stress responses. However, the understanding of KCS genes in maize remains limited. In this study, we present a comprehensive analysis of ZmKCS genes, identifying 29 KCS genes that are unevenly distributed across nine maize chromosomes through bioinformatics approaches. These ZmKCS proteins varied in length and molecular weight, suggesting functional diversity. Phylogenetic analysis categorized 182 KCS proteins from seven species into six subgroups, with maize showing a closer evolutionary relationship to other monocots. Collinearity analysis revealed 102 gene pairs between maize and three other monocots, whereas only five gene pairs were identified between maize and three dicots, underscoring the evolutionary divergence of KCS genes between monocotyledonous and dicotyledonous plants. Structural analysis revealed that 20 out of 29 ZmKCS genes are intronless. Subcellular localization prediction and experimental validation suggest that most ZmKCS proteins are likely localized at the plasma membrane, with some also present in mitochondria and chloroplasts. Analysis of the cis-acting elements within the ZmKCS promoters suggested their potential involvement in abiotic stress responses. Notably, expression analysis under abiotic stresses highlighted ZmKCS17 as a potential key gene in the stress response of maize, which presented an over 10-fold decrease in expression under salt and drought stresses within 48 h. This study provides a fundamental understanding of ZmKCS genes, paving the way for further functional characterization and their potential application in maize breeding for enhanced stress tolerance.

Application of spatial metrology models in cell molecular localization and functional prediction

Understanding a protein’s exact cellular location is often essential to understanding its function. Even with the advancements in computer approaches, protein localization prediction indeed faces major obstacles such as interpretability and handling numerous localization sites. In this research, a novel approach, Squirrel Search Optimized Dynamic Visual Geometry Group Network (SSO-DVGG), is proposed to improve protein sub-cellular localization predictions by utilizing spatial metrology models to tackle these problems. With its simplified architecture, SSO-DVGG can explain whether a protein is directed to particular cellular sites, as well as identify important sequence components like sorting motifs or localization signals. This model allows users to select acceptable error levels by providing a confidence estimate for each prediction and highlighting sequence properties that are responsible for localization. This makes the model interpretable. Furthermore, SSO-DVGG uses a probabilistic methodology and integrates a large amount of data from dual-targeted proteins, which enables it to predict multiple localization locations per protein accurately. SSO-DVGG outperforms the best predictors and shows superior capacity to predict multiple localizations when tested on several independent datasets. By providing a clear and accurate understanding of protein distribution and function, this method promotes the application of spatial metrology models in cell molecular localization and functional prediction.

Mitochondrial aspartate aminotransferase (GOT2) protein as a potential cryodamage biomarker in rooster spermatozoa cryopreservation.

Spermatozoa cryopreservation has been widely used for animal genetic conservation. Despite advances in chicken semen cryopreservation, the mechanism of spermatozoa cryodamage remains to be revealed. The cryopreservation process induces motion parameter decreased, structure damaged, proteomic and antioxidant system remodeled in spermatozoa. Mitochondrial glutamate-oxaloacetate transaminase 2 (GOT2) is part of the malate aspartate shuttle, which is ubiquitous in mitochondria and is associated with cellular metabolism regulation. Thus, this study is the first to investigate GOT2 biological role in chicken spermatozoa during freezing process. The results showed that the sperm total motility, straight-line velocity (VSL) and mitochondrial membrane potential (MMP) of the frozen group were significantly lower than these of the fresh group (P < 0.05). The fresh sperm mitochondrial membrane was continuous and mitochondrial matrix was dense and homogeneous. However, after the freezing-thawing, the density of the matrix was reduced, and the mitochondria appeared slightly swollen and membrane damaged. In chicken sperm, the GOT2 protein was localized in the head and the midpiece of spermatozoa where mitochondria are located by immunostaining analysis. This was consistent with the subcellular localization prediction. GOT2 protein was more abundant in the fresh sperm than in the frozen sperm, which indicated that freezing may lead to sperm mitochondrial damage, reduced GOT protein expression, and affected sperm motility and fertility. The protein-protein interaction prediction of GOT2 protein indicated that its ten most confident interactors were predominantly mitochondria-related proteins. The binding ability was higher between GOT2 protein and two mitochondria-targeted antioxidants, SkQ1 and Mito-TEMPO, respectively. In conclusion, GOT2 played an important role in chicken spermatozoa, which was possibly associated with the regulation of mitochondria function and spermatozoa metabolism. Moreover, it may be a potential cryodamage improvement target for spermatozoa. However, the underlying mechanism of GOT2 in spermatozoa cryopreservation needs further exploration.

Genome-wide survey of chlorophyllase (CLH) gene family in seven Rosaceae and functional characterization of MdCLH1 in apple (Malus domestica) leaf photosynthesis

Chlorophyll, a crucial pigment in plant photosynthesis, undergoes dynamic synthesis and degradation processes throughout the growth cycle of plants. Chlorophyllase (CLH) plays a crucial role in the degradation of chlorophyll by removing the phytol group from chlorophyll A to produce chlorophyllide A. However, Rosaceae species remain underexplored in terms of understanding the functional divergences among CLH gene family members (CLHs) involved in chlorophyll catabolism and photosynthesis. The apple (Malus domestica) CLHs also requires further systematic characterization and identification. In this study, 20 CLHs (MdCLH1-4, FvCLH1-2, PpCLH1-2, PcCLH1-3, PaCLH1-2, RrCLH1-4, RcCLH1-3) were identified from seven species belonging to the Rosaceae family. The chromosomal distribution of these gene family members is mostly separate across all species, except in Rosa rugosa. The number of amino acids encoded by these genes ranges from 171 aa to 391 aa, possessing a theoretical isoelectric point (PI) of 5.46-9.59, and a relative molecular weight of 18313.07D to 42413.21D. Secondary structure predictions highlight α-helix and random coil conformations as the dominant structural elements of CLH proteins present in Rosaceae species. Subcellular localization predictions indicate that all CLH proteins are expressed in chloroplasts, while MdCLH4 is uniquely localized to the nucleus. Phylogenetic analysis reveals high homology and close evolutionary relationships among the genes in three subfamilies. All these 20 CLHs contain elements responsive to phytohormones, environmental stress, and light. Furthermore, transcriptomic profiling using geneChip expression array coupled with qRT-PCR analyses revealed a heightened transcriptional activity of MdCLHs in leaf tissues and protective tissue of annual shoots as compared to other plant components. Additional experimental evidence specifically indicates MdCLH1 is located in the chloroplasts of tobacco leaves. Notably, MdCLH1 transient expression in apple leaves decreased chlorophyll a, carotenoids, total chlorophyll content, non - photochemical quenching coefficient (NPQ), and intercellular CO₂ concentration (Ci), while increasing chlorophyll b content, effective PSII quantum yield [Y(II)], net photosynthetic rate (Pn), stomatal conductance (Gs), and electron transport rate (ETR). These suggest MdCLH1 enhances light energy conversion in PSII by modulating chlorophyll degradation, potentially improving photosynthetic efficiency and reducing the potential for photoinhibition. This study lays a solid foundation for further exploration into the functional roles of CLHs in the Rosaceae family.

Predicting the subcellular location of prokaryotic proteins with DeepLocPro.

Protein subcellular location prediction is a widely explored task in bioinformatics because of its importance in proteomics research. We propose DeepLocPro, an extension to the popular method DeepLoc, tailored specifically to archaeal and bacterial organisms. DeepLocPro is a multiclass subcellular location prediction tool for prokaryotic proteins, trained on experimentally verified data curated from UniProt and PSORTdb. DeepLocPro compares favorably to the PSORTb 3.0 ensemble method, surpassing its performance across multiple metrics in our benchmark experiment. The DeepLocPro prediction tool is available online at https://ku.biolib.com/deeplocpro and https://services.healthtech.dtu.dk/services/DeepLocPro-1.0/.

MYB transcription factors in Peucedanum Praeruptorum Dunn: the diverse roles of the R2R3-MYB subfamily in mediating coumarin biosynthesis

BackgroundThe MYB superfamily (v-myb avian myeloblastosis viral oncogene homolog) plays a role in plant growth and development, environmental stress defense, and synthesis of secondary metabolites. Little is known about the regulatory function of MYB genes in Peucedanum praeruptorum Dunn, although many MYB family members, especially R2R3-MYB genes, have been extensively studied in model plants.ResultsA total of 157 R2R3-MYB transcription factors from P. praeruptorum were identified using bioinformatics analysis. Comprehensive analyses including chromosome location, microsynteny, gene structure, conserved motif, phylogenetic tree, and conserved domain were further performed. The length of the 157 transcription factors ranged from 120 to 1,688 amino acids (molecular weight between 14.21 and 182.69 kDa). All proteins were hydrophilic. Subcellular localization predictions showed that 155 PpMYB proteins were localized in the nucleus, with PpMYB12 and PpMYB157 localized in the chloroplasts and mitochondria, respectively. Ten conserved motifs were identified in the PpMYBs, all of which contained typical MYB domains. Transcriptome analysis identified 47,902 unigenes. Kyoto Encyclopedia of Genes and Genomes analysis revealed 136 pathways, of which 524 genes were associated with the phenylpropanoid pathway. Differential expressed genes (DEGs) before and after bolting showed that 11 genes were enriched in the phenylpropanoid pathway. Moreover, the expression patterns of transcription genes were further verified by qRT-PCR. With high-performance liquid chromatography (HPLC), 8 coumarins were quantified from the root, stem, and leaf tissue samples of P. praeruptorum at different stages. Praeruptorin A was found in both roots and leaves before bolting, whereas praeruptorin B was mainly concentrated in the roots, and the content of both decreased in the roots and stems after bolting. Praeruptorin E content was highest in the leaves and increased with plant growth. The correlation analysis between transcription factors and coumarin content showed that the expression patterns of PpMYB3 and PpMYB103 in roots align with the accumulation trends of praeruptorin A, praeruptorin B, praeruptorin E, scopoletin, and isoscopoletin, which declined in content after bolting, suggesting that these genes may positively regulate the biosynthesis of coumarins. Eleven distinct metabolites and 48 DEGs were identified. Correlation analysis revealed that the expression of all DEGs were significantly related to the accumulation of coumarin metabolites, indicating that these genes are involved in the regulation of coumarin biosynthesis.ConclusionsR2R3-MYB transcription factors may be involved in the synthesis of coumarin. Our findings provide basic data and a rationale for future an in-depth studies on the role of R2R3-MYB transcription factors in the growth and regulation of coumarin synthesis.

Ectopic expression of HvbHLH132 from hulless barley reduces cold tolerance in transgenic Arabidopsis thaliana.

Overexpression of HvbHLH132 from hulless barley impairs in chilling and freezing tolerance at the seedlings stage in Arabidopsis thaliana The basic helix-loop-helix (bHLH) transcription factors (TF) are ubiquitously existed in eukaryote and play crucial roles in numerous biological processes. However, the characterization of their members and functions in hulless barley remains limited. Here, we conducted a genome-wide identification of the HvbHLH gene family and assessed the role of HvbHLH132 in cold stress tolerance. We identified 141 HvbHLH genes, which were categorized into twelve subfamilies. Subcellular localization predictions indicated that the majority of HvbHLH proteins were localized in the nucleus. cis-Acting element analysis revealed that the promoter regions of the HvbHLH family contain diverse elements associated with various biological processes. Expression profiling of the 141 HvbHLH genes in two extreme varieties revealed that HvbHLH132 was significantly induced and exhibited substantial differential expression under cold stress. Analyses of subcellular localization and transactivation activity confirmed that HvbHLH132 specifically localized in the nucleus and contributed to transcriptional activation. Furthermore, overexpression of HvbHLH132 in Arabidopsis resulted in impaired chilling and freezing tolerance at the seedling stage, leading to biochemical changes unfavorable for freezing stress. Additionally, the expression of some cold-responsive genes (COR) genes was significantly less induced compared to wild type under freezing stress. This study provides comprehensive insight into the HvbHLH gene family and reveals a critical role of HvbHLH132 in regulating cold tolerance in plants.

Functions of MdTINY, a member of the apple dehydration responsive element binding-A4

The dehydration responsive element binding (DREB) transcription factors play an important role in plant growth and development and are extensively involved in plant responses to abiotic stress. The DREB family contains six subfamilies, and TINY belongs to the DREB-A4 subfamily. The Arabidopsis thaliana TINY gene, AtTINY, plays a role in regulating plant growth and responses to stress. In order to investigate the evolutionary characterization of the DREB-A4 subfamily and the biological function of the MdTINY gene in apple (Malus domestica), in this study, we used the databases GDDH13 and TAIR and online tools (Expasy and WoLF PSORT) to study the biological information of the DREB-A4 subfamily in apple. In addition, the tertiary structures of the proteins were predicted. The apple DREB-A4 subfamily contained 22 genes, all of which had a conserved AP2 domain, and subcellular localization predictions showed that DREB-A4 subfamily proteins were mainly located in the nucleus. The transgenic calli of MdTINY were obtained by the Agrobacterium-mediated transformation method, and the main biological functions of MdTINY were explored by quantitative real-time PCR (qRT-PCR) combined with anthocyanin content determination. MdTINY shared the highest amino acid sequence similarity with AtTINY. The coding region of MdTINY had a full length of 759 bp, encoding 252 amino acid residues. Analysis of the promoter elements and expression patterns indicated that MdTINY was responsive to light and multiple stress conditions. MdTINY was localized in the nucleus and had transcriptional autoactivation activity. The overexpression of MdTINY in calli inhibited normal growth and promoted anthocyanoside accumulation. These results indicated that MdTINY negatively regulated apple plant growth and promoted fruit coloring, providing a candidate gene for the breeding of apple varieties with high quality of fruit color.

Impact of Alignments on the Accuracy of Protein Subcellular Localization Predictions.

Alignments in bioinformatics refer to the arrangement of sequences to identify regions of similarity that can indicate functional, structural, or evolutionary relationships. They are crucial for bioinformaticians as they enable accurate predictions and analyses in various applications, including protein subcellular localization. The predictive model used in this article is based on a deep - convolutional architecture. We tested configurations of Deep N-to-1 convolutional neural networks of various depths and widths during experimentation for the evaluation of better-performing values across a diverse set of eight classes. For without alignment assessment, sequences are encoded using one-hot encoding, converting each character into a numerical representation, which is straightforward for non-numerical data and useful for machine learning models. For with alignments assessment, multiple sequence alignments (MSAs) are created using PSI-BLAST, capturing evolutionary information by calculating frequencies of residues and gaps. The average difference in peak performance between models with alignments and without alignments is approximately 15.82%. The average difference in the highest accuracy achieved with alignments compared with without alignments is approximately 15.16%. Thus, extensive experimentation indicates that higher alignment accuracy implies a more reliable model and improved prediction accuracy, which can be trusted to deliver consistent performance across different layers and classes of subcellular localization predictions. This research provides valuable insights into prediction accuracies with and without alignments, offering bioinformaticians an effective tool for better understanding while potentially reducing the need for extensive experimental validations. The source code and datasets are available at http://distilldeep.ucd.ie/SCL8/.

The improved de Bruijn graph for multitask learning: predicting functions, subcellular localization, and interactions of noncoding RNAs.

Noncoding RNA refers to RNA that does not encode proteins. The lncRNA and miRNA it contains play crucial regulatory roles in organisms, and their aberrant expression is closely related to various diseases. Traditional experimental methods for validating the interactions of these RNAs have limitations, and existing prediction models exhibit relatively limited functionality, relying on isolated feature extraction and performing poorly in handling various types of small sample tasks. This paper proposes an improved de Bruijn graph that can inject RNA structural information into the graph while preserving sequence information. Furthermore, the improved de Bruijn graph enables graph neural networks to learn broader dependencies and correlations among data by introducing richer edge relationships. Meanwhile, the multitask learning model, DVMnet, proposed in this paper can handle multiple related tasks, and we optimize model parameters by integrating the total loss of three tasks. This enables multitask prediction of RNA interactions, disease associations, and subcellular localization. Compared with the best existing models in this field, DVMnet has achieved the best performance with a 3% improvement in the area under the curve value and demonstrates robust results in predicting diseases and subcellular localization. The improved de Bruijn graph is also applicable to various scenarios and can unify the sequence and structural information of various nucleic acids into a single graph.

Genome-wide identification and expression analysis of phosphate-sensing SPX proteins in oats.

Phosphorus is indispensable to plant growth and development. Soil phosphorus deficiency poses a substantial constraint on crop yield. SPXs play pivotal roles in phosphate transport and absorption in plants. Yet, the functions of SPXs of oat (Avena sativa L.) under abiotic stresses remain unclear. In this study, we conducted a genome-wide analysis of 169 SPXs from hexaploid oat and five closely related plant species. All homologous AsSPXs were found to arise from duplication events and depict a strong purifying selection. Subcellular localization prediction revealed that AsSPXs were mainly located on the plasma membrane. Seventeen cis-acting elements, predominantly comprising light-, low temperature-, abscisic acid-, and drought-responsive elements, were dispersed in the promoter regions of AsSPXs. Analysis of cis-regulatory elements, protein-protein interaction networks, and qRT-PCR showed that AsSPXs are not solely involved in phosphorus starvation response but also in various stress responses. Notably, AsSPX18-5D (AVESA.00001b.r3.5Dg0002895) exerted pivotal roles in conferring resistance against low phosphorus, salt, and ABA treatments. Our study aimed to explore important stress-resistant genes in oat. Our results could provide a basis for future studies on the evolution and functions of the AsSPX gene family and a crucial foundation for comprehending how oat responds to environmental stresses.

Genome-Wide Identification and Analysis of WD40 Family and Its Expression in F. vesca at Different Coloring Stages.

WD40 proteins play important roles in the synthesis and regulation of anthocyanin, the regulation of plant morphology and development, and the response to various abiotic stresses. However, the role of WD40 in Fragaria vesca (F. vesca) has not been studied. In this study, a total of 216 FvWD40 family members were identified, which were divided into four subfamilies based on evolutionary tree analysis. Subcellular localization predictions show that FvWD40 family members are mainly localized in chloroplasts, nuclei, and cytoplasm. An analysis of collinearity revealed a total of eight pairs of intraspecific collinearity of the FvWD40 gene family, and interspecific collinearity showed that the FvWD40 gene family covaried more gene pairs with Arabidopsis thaliana (Arabidopsis) than with rice (Oryza sativa). Promoter cis-acting elements revealed that the FvWD40 gene family contains predominantly light, hormone, and abiotic stress response elements. Tissue-specific expression analysis showed that a number of members including FvWD40-111 and FvWD40-137 were highly expressed in all tissues, and a number or members including FvWD40-97 and FvWD40-102 were lowly expressed in all tissues. The FvWD40 gene family was found to be expressed at all four different coloring stages of F. vesca by qRT-PCR, with lower expression at the 50% coloring stage (S3). FvWD40-24, FvWD40-50, and FvWD40-60 showed the highest expression during the white fruit stage (S1) period, suggesting that these genes play a potential regulatory role in the pre-fruit coloring stage. FvWD40-62, FvWD40-88 and FvWD40-103 had the highest expression at the 20% coloration stage (S2), and FvWD40-115, FvWD40-170, FvWD40-184 and FvWD40-195 had the highest expression at the full coloration stage (S4). These results suggest a potential role for these genes during fruit coloration. This study lays a foundation for further research on the function of the WD40 gene family.

Deciphering Membrane Proteins Through Deep Learning Models by Revealing Their Locale Within the Cell.

Membrane proteins constitute essential biomolecules attached to or integrated into cellular and organelle membranes, playing diverse roles in cellular processes. Their precise localization is crucial for understanding their functions. Existing protein subcellular localization predictors are predominantly trained on globular proteins; their performance diminishes for membrane proteins, explicitly via deep learning models. To address this challenge, the proposed study segregates membrane proteins into three distinct locations, including the plasma membrane, internal membrane, and membrane of the organelle, using deep learning algorithms including recurrent neural networks (RNN) and Long Short-Term Memory (LSTM). A redundancy-curtailed dataset of 3000 proteins from the MemLoci approach is selected for the investigation, along with incorporating pseudo amino acid composition (PseAAC). PseAAC is an exemplary technique for extracting protein information hidden in the amino acid sequences. After extensive testing, the results show that the accuracy for LSTM and RNN is 83.4% and 80.5%, respectively. The results show that the LSTM model outperforms the RNN and is most commonly employed in proteomics.

Natural Language Prompts Guide the Design of Novel Functional Protein Sequences.

The advent of natural language interaction with machines has ushered in new innovations in text-guided generation of images, audio, video, and more. In this arena, we introduce Bio logical M ulti- M odal M odel ( BioM3 ), as a novel framework for designing functional proteins via natural language prompts. This framework integrates natural language with protein design through a three-stage process: aligning protein and text representations in a joint embedding space learned using contrastive learning, refinement of the text embeddings, and conditional generation of protein sequences via a discrete autoregressive diffusion model. BioM3 synthe-sizes protein sequences with detailed descriptions of the protein structure, lineage, and function from text annotations to enable the conditional generation of novel sequences with desired attributes through natural language prompts. We present in silico validation of the model predictions for subcellular localization prediction, reaction classification, remote homology detection, scaffold in-painting, and structural plausibility, and in vivo and in vitro experimental tests of natural language prompt-designed synthetic analogs of Src-homology 3 (SH3) domain proteins that mediate signaling in the Sho1 osmotic stress response pathway in baker's yeast. BioM3 possesses state-of-the-art performance in zero-shot prediction and homology detection tasks, and generates proteins with native-like tertiary folds and wild-type levels of experimentally assayed function.

Genome-wide characterization of melatonin biosynthetic pathway genes in carnation (Dianthus caryophyllus L.) and their expression analysis in response to exogenous melatonin

Carnation (Dianthus caryophyllus L.) is a prominent floricultural crop valued for its diverse colors, offering significant economic and ornamental value globally. However, the global demand for its flowers makes flower yield an important attribute that relies on the quantity of lateral branches in the crop. Melatonin as a multi-regulatory phytohormone play vital functions in governing plant growth and development. It is synthesized from tryptophan via four key enzymes. Tryptophan decarboxylase (TDC), Tryptamine 5-hydroxylase (T5H), Serotonin N-acetyltransferase (SNAT), and N-Acetylserotonin O-methyltransferase (ASMT). Although the significance of melatonin is recognized, its impact on the growth and development of carnation remains understudied. In the current study, we investigated the effect of exogenous melatonin at different concentrations, on growth pattern of carnation, followed by genome-wide characterization, in-silico analysis and expression profiling of melatonin biosynthetic pathway genes. Results showed increased branching and reduced height with increased melatonin concentrations up to a point. In-silico analysis identified ten genes in the melatonin biosynthetic pathway, including two TDC, two T5H, one SNAT, and five ASMT members. Domain analysis confirmed the presence of characteristic domains such as pyridoxal-dependent decarboxylase, cytochrome P450, Acetyltransferase_1, and O-methyltransferase. Physiochemical properties, gene structure, conserved motifs, promoter regions, gene ontology, synteny, and evolutionary relationships through phylogeny were also analysed. Sub-cellular localization predictions showed distribution of melatonin biosynthetic enzymes across various cellular compartments. Expression analysis of these genes under different exogenous melatonin concentrations (100, 200, 300, 400, 500, and 1000 µM) revealed significant upregulation at 100 µM and 500 µM, while no change was observed at 1000 µM. These findings suggest that optimal exogenous melatonin concentrations enhance the expression of biosynthetic pathway genes ultimately led to increased branching in carnation due to increased endogenous melatonin levels. This study establishes a basis for future functional characterization of melatonin biosynthetic pathway genes to elucidate their roles in carnation growth and development.

Comparative analysis of the NF-Y transcription factor family identifies VaNF-YA6 as a positive regulator of salt and drought tolerance in grapevine

Nuclear factor Y transcription factors (NF-Y TFs) play crucial roles in plant responses to abiotic stress. However, there is a lack of research on the comparative analysis of evolutionary relationships, real-time quantitative fluorescence PCR (RT-qPCR), and functions of NF-Y TFs to screen key NF-Y TFs that are resistant to salt and drought stresses between Vitis vinifera (V. vinifera) and Vitis amurensis (V. amurensis). In this study, 27 and 26 NF-Y TFs were identified in V. vinifera and V. amurensis, respectively, and were divided into three subgroups. Subcellular localization prediction revealed that NF-Ys TFs were mainly located in the nucleus. Interestingly, the NF-YA protein sequence of ‘NTKKLDWEFWGCCDDCEKWFGGCC’ was lost in the V. vinifera compared to V. amurensis, whereas the sequence ‘SSVYSQPWWGHSIVCVA’ was gained, thus, these sequences might be closely related to the functions performed. RT-qPCR analysis of ‘Pinot Noir’ (cultivated variety) and ‘Zuoyouhong’ (wild variety) plantlets demonstrated that the expression levels of VaNF-YA6, VaNF-YB5, VvNF-YA3, VvNF-YA5, and VvNF-YC2 were significantly upregulated under 400 mmol·L-1 NaCl and 10% PEG treatments for 24 h Subcellular localization showed that the VaNF-YA6-GFP fusion protein was functioned primarily in the nucleus. Overexpression of VaNF-YA6 in grapevine leaves and Arabidopsis thaliana (Arabidopsis) could significantly enhance tolerance to salt and drought stresses by improving VvSOS2, VvSOS3, VvABF3, VvCPK6 expression levels, enzyme activities, and other protective substances. In summary, our study provides a theoretical basis for the further use of VaNF-YA6 to improve salt and drought resistance in grapevines.

MMLmiRLocNet: miRNA Subcellular Localization Prediction based on Multi-view Multi-label Learning for Drug Design.

Identifying subcellular localization of microRNAs (miRNAs) is essential for comprehensive understanding of cellular function and has significant implications for drug design. In the past, several computational methods for miRNA subcellular localization is being used for uncovering multiple facets of RNA function to facilitate the biological applications. Unfortunately, most existing classification methods rely on a single sequencebased view, making the effective fusion of data from multiple heterogeneous networks a primary challenge. Inspired by multi-view multi-label learning strategy, we propose a computational method, named MMLmiRLocNet, for predicting the subcellular localizations of miRNAs. The MMLmiRLocNet predictor extracts multi-perspective sequence representations by analyzing lexical, syntactic, and semantic aspects of biological sequences. Specifically, it integrates lexical attributes derived from k-mer physicochemical profiles, syntactic characteristics obtained via word2vec embeddings, and semantic representations generated by pre-trained feature embeddings. Finally, module for extracting multi-view consensus-level features and specific-level features was constructed to capture consensus and specific features from various perspectives. The full connection networks are utilized as the output module to predict the miRNA subcellular localization. Experimental results suggest that MMLmiRLocNet outperforms existing methods in terms of F1, subACC, and Accuracy, and achieves best performance with the help of multi-view consensus features and specific features extract network.