RESUMO
Language models pretrained by self-supervised learning (SSL) have been widely utilized to study protein sequences, while few models were developed for genomic sequences and were limited to single species. Due to the lack of genomes from different species, these models cannot effectively leverage evolutionary information. In this study, we have developed SpliceBERT, a language model pretrained on primary ribonucleic acids (RNA) sequences from 72 vertebrates by masked language modeling, and applied it to sequence-based modeling of RNA splicing. Pretraining SpliceBERT on diverse species enables effective identification of evolutionarily conserved elements. Meanwhile, the learned hidden states and attention weights can characterize the biological properties of splice sites. As a result, SpliceBERT was shown effective on several downstream tasks: zero-shot prediction of variant effects on splicing, prediction of branchpoints in humans, and cross-species prediction of splice sites. Our study highlighted the importance of pretraining genomic language models on a diverse range of species and suggested that SSL is a promising approach to enhance our understanding of the regulatory logic underlying genomic sequences.
Assuntos
Splicing de RNA , Vertebrados , Animais , Humanos , Sequência de Bases , Vertebrados/genética , RNA , Aprendizado de Máquina SupervisionadoRESUMO
Functional peptides play crucial roles in various biological processes and hold significant potential in many fields such as drug discovery and biotechnology. Accurately predicting the functions of peptides is essential for understanding their diverse effects and designing peptide-based therapeutics. Here, we propose CELA-MFP, a deep learning framework that incorporates feature Contrastive Enhancement and Label Adaptation for predicting Multi-Functional therapeutic Peptides. CELA-MFP utilizes a protein language model (pLM) to extract features from peptide sequences, which are then fed into a Transformer decoder for function prediction, effectively modeling correlations between different functions. To enhance the representation of each peptide sequence, contrastive learning is employed during training. Experimental results demonstrate that CELA-MFP outperforms state-of-the-art methods on most evaluation metrics for two widely used datasets, MFBP and MFTP. The interpretability of CELA-MFP is demonstrated by visualizing attention patterns in pLM and Transformer decoder. Finally, a user-friendly online server for predicting multi-functional peptides is established as the implementation of the proposed CELA-MFP and can be freely accessed at http://dreamai.cmii.online/CELA-MFP.
Assuntos
Aprendizado Profundo , Peptídeos , Peptídeos/química , Biologia Computacional/métodos , Software , Humanos , Algoritmos , Bases de Dados de ProteínasRESUMO
BACKGROUND: With the improvements in laparoscopic or robotic surgical techniques and instruments, a growing number of surgeons have attempted to complete all digestive tract reconstruction intracorporeally; these procedures include totally robotic gastrectomy (TRG) and totally laparoscopic gastrectomy (TLG). This study aimed to evaluate the safety and feasibility of the TRG and compare the short-term outcomes of the TRG and TLG in patients with gastric cancer. METHODS: Between January 2018 and June 2023, 346 consecutive patients who underwent TRG or TLG at a high-volume academic gastric cancer specialty center were included. 1:1 propensity score matching (PSM) was performed to reduce confounding bias. The surgical outcomes, postoperative morbidity, and surgical burden were compared in PSM cohort. RESULTS: After PSM, a well-balanced cohort of 194 patients (97 in each group) was included in the analysis. The total operation time of the TRG group was significantly longer than that of the TLG group (244.9 vs. 213.0 min, P < 0.001). There was no significant difference in the effective operation time between the 2 groups (217.8 vs. 207.2 min, P = 0.059). The digestive tract reconstruction time of the TRG group was significantly shorter than that of the TLG group (39.4 vs. 46.7 min, P < 0.001). The mean blood loss in the TRG group was less than that in the TLG group (101.1 vs. 126.8 mL, P = 0.014). The TRG group had more retrieved lymph nodes in the suprapancreatic area than that in the TLG group (16.6 vs 14.2, P = 0.002). The TRG group had a lower surgery task load index (38.9 vs. 43.1, P < 0.001) than the TLG group. No significant difference was found in terms of postoperative morbidity between the 2 groups (14.4% vs. 16.5%, P = 0.691). CONCLUSION: This study demonstrated that TRG is a safe and feasible procedure, and is preferable to TLG in terms of invasion and ergonomics. The TRG may maximize the superiority of robotic surgical systems and embodies the theory of minimally invasive surgery.
Assuntos
Gastrectomia , Laparoscopia , Duração da Cirurgia , Pontuação de Propensão , Procedimentos Cirúrgicos Robóticos , Neoplasias Gástricas , Humanos , Neoplasias Gástricas/cirurgia , Gastrectomia/métodos , Procedimentos Cirúrgicos Robóticos/métodos , Laparoscopia/métodos , Masculino , Feminino , Pessoa de Meia-Idade , Idoso , Estudos Retrospectivos , Estudos de Viabilidade , Resultado do Tratamento , Complicações Pós-Operatórias/epidemiologia , Complicações Pós-Operatórias/etiologiaRESUMO
Mining the potential of traditional Chinese medicine (TCM) in treating modern diseases requires a profound understanding of its action mechanism and a comprehensive knowledge system that seamlessly bridges modern medical insights with traditional theories. However, existing databases for modernizing TCM are plagued by varying degrees of information loss, which impede the multidimensional dissection of pharmacological effects. To address this challenge, we introduce traditional Chinese medicine modernization (TCMM), the currently largest modernized TCM database that integrates pioneering intelligent pipelines. By aligning high-quality TCM and modern medicine data, TCMM boasts the most extensive TCM modernization knowledge, including 20 types of modernized TCM concepts such as prescription, ingredient, target and 46 biological relations among them, totaling 3,447,023 records. We demonstrate the efficacy and reliability of TCMM with two features, prescription generation and knowledge discovery, the outcomes show consistency with biological experimental results. A publicly available web interface is at https://www.tcmm.net.cn/.
RESUMO
Large language models have made significant strides in natural language processing, enabling innovative applications in molecular science by processing textual representations of molecules. However, most existing language models cannot capture the rich information with complex molecular structures or images. In this paper, we introduce GIT-Mol, a multi-modal large language model that integrates the Graph, Image, and Text information. To facilitate the integration of multi-modal molecular data, we propose GIT-Former, a novel architecture that is capable of aligning all modalities into a unified latent space. We achieve a 5%-10% accuracy increase in properties prediction and a 20.2% boost in molecule generation validity compared to the baselines. With the any-to-language molecular translation strategy, our model has the potential to perform more downstream tasks, such as compound name recognition and chemical reaction prediction.
Assuntos
Idioma , Processamento de Linguagem NaturalRESUMO
Cytokines are small protein molecules that exhibit potent immunoregulatory properties, which are known as the essential components of the tumor immune microenvironment (TIME). While some cytokines are known to be universally upregulated in TIME, the unique cytokine expression patterns have not been fully resolved in specific types of cancers. To address this challenge, we develop a TIME single-cell RNA sequencing (scRNA-seq) dataset, which is designed to study cytokine expression patterns for precise cancer classification. The dataset, including 39 cancers, is constructed by integrating 684 tumor scRNA-seq samples from multiple public repositories. After screening and processing, the dataset retains only the expression data of immune cells. With a machine learning classification model, unique cytokine expression patterns are identified for various cancer categories and pioneering applied to cancer classification with an accuracy rate of 78.01%. Our method will not only boost the understanding of cancer-type-specific immune modulations in TIME but also serve as a crucial reference for future diagnostic and therapeutic research in cancer immunity.
Assuntos
Citocinas , Neoplasias , Humanos , Citocinas/genética , Neoplasias/diagnóstico , Neoplasias/genética , Análise de Sequência de RNA/métodos , Aprendizado de Máquina , Análise de Célula Única/métodos , Microambiente TumoralRESUMO
The HLA-DRB1*16:76 allele differs from HLA-DRB1*16:02:01 by one nucleotide substitution (A > G) at position 37 in exon 1.
Assuntos
Cadeias HLA-DRB1 , Humanos , Cadeias HLA-DRB1/genética , Sequência de Bases , Alelos , Éxons/genética , ChinaRESUMO
The HLA-A*11:01:124 allele differs from HLA-A*11:01:01 by one nucleotide substitution, (C > T) position 459 in exon 3.
Assuntos
Antígenos HLA-A , Humanos , Alelos , China , Éxons/genética , Antígenos HLA-A/genética , População do Leste AsiáticoRESUMO
Self-assembled monolayers (SAMs) have shown great potential as hole injection materials for perovskite light-emitting diodes due to their low parasitic absorption and ability to adjust energy level alignment. However, the head and anchoring groups on SAM molecules with significant differences in polarity can lead to the formation of micelles in the commonly used alcoholic processing solvent, inhibiting the formation of an intact SAM. In this work, the introduction of methyl groups on carbazole in the phosphonic-acid-based SAM materials is found to facilitate energy level alignment and promote the formation of compact SAMs. The alternative molecular structure also enhances the solvent resistance of poly(9-vinylcarbazole), suppressing interfacial defect densities and nonradiative recombination processes in the emissive perovskites. PeLEDs based on the methyl-containing SAMs exhibit â¼30% enhancement in efficiency. These findings contribute to a better understanding of the design of SAM materials for PeLED applications.
RESUMO
Perovskite LEDs (PeLEDs) have emerged as a next-generation light-emitting technology. Recent breakthroughs were made in achieving highly stable near-infrared and green PeLEDs. However, the operational lifetimes (T50) of visible PeLEDs under high current densities (>10 mA cm-2) remain unsatisfactory (normally <100 h), limiting the possibilities in solid-state lighting and AR/VR applications. This problem becomes more pronounced for mixed-halide (e.g., red and blue) perovskite emitters in which critical challenges such as halide segregation and spectral instability are present. Here, we demonstrate bright and stable red PeLEDs based on mixed-halide perovskites, showing measured T50 lifetimes of up to â¼357 h at currents of ≥25 mA cm-2, a record for the operational stability of visible PeLEDs under high current densities. The devices produce intense and stable emission with a maximum luminance of 28,870 cd m-2 (radiance: 1584 W sr-1 m-2), which is record-high for red PeLEDs. Key to this demonstration is the introduction of sulfonamide, a dipolar molecular stabilizer that effectively interacts with the ionic species in the perovskite emitters. It suppresses halide segregation and migration into the charge-transport layers, resulting in enhanced stability and brightness of the mixed-halide PeLEDs. These results represent a substantial step toward bright and stable PeLEDs for emerging applications.