RESUMO
Cancer of unknown primary (CUP) represents metastatic cancer where the primary site remains unidentified despite standard diagnostic procedures. To determine the tumor origin in such cases, we developed BPformer, a deep learning method integrating the transformer model with prior knowledge of biological pathways. Trained on transcriptomes from 10,410 primary tumors across 32 cancer types, BPformer achieved remarkable accuracy rates of 94%, 92%, and 89% in primary tumors and primary and metastatic sites of metastatic tumors, respectively, surpassing existing methods. Additionally, BPformer was validated in a retrospective study, demonstrating consistency with tumor sites diagnosed through immunohistochemistry and histopathology. Furthermore, BPformer was able to rank pathways based on their contribution to tumor origin identification, which helped to classify oncogenic signaling pathways into those that are highly conservative among different cancers versus those that are highly variable depending on their origins.
Assuntos
Neoplasias Primárias Desconhecidas , Humanos , Neoplasias Primárias Desconhecidas/genética , Neoplasias Primárias Desconhecidas/patologia , Neoplasias Primárias Desconhecidas/metabolismo , Neoplasias Primárias Desconhecidas/diagnóstico , Transdução de Sinais/genética , Transcriptoma , Aprendizado Profundo , Estudos RetrospectivosRESUMO
Dual-mode signal output platforms have demonstrated considerable promise due to their improved anti-interference capability and inherent signal self-correction. Nevertheless, traditional discrete-distributed signal probes often encounter significant drawbacks, including limited mass transfer efficiency, diminished signal strength, and instability in intricate biochemical environments. In response to these challenges, a scalable and hyper-compacted 3D DNA nanoplatform resembling "periodic focusing heliostat" has been developed for synergistically enhanced fluorescence (FL) and surface-enhanced Raman spectroscopy (SERS) biosensing of miRNA in cancer cells. Our approach utilized a distinctive assembly strategy integrating gold nanostars (GNS) as fundamental "heliostat units" linked by palindromic DNA sequences to facilitate each other hand-in-hand cascade alignment and condensed into large scale nanostructures. This configuration was further augmented by the incorporation of gold nanoparticles (GNP) via strong Au-S bonds, resulting in a sturdy framework for improved signal transduction. The initiation of this assembly process was mediated by the hybridization of dsDNA to miRNA-21, which served as a primer for polymerization and nicking reactions, thus generating a multifunctional T2 probe. This probe is intricately designed with three distinct parts: a 3'-palindromic end for structural integrity, a central region for capturing SERS-active probes (Cy3-P2), and a 5'-segment for attaching fluorescence reporters. Upon integration T2 into the GNS-based heliostat unit, it promotes palindromic arm-induced aggregation and plasma exciton coupling between plasma nanoparticles and signal transduction tags. This clustered arrangement creates a high-density "hot spot" array that maximizes the local electromagnetic fields necessary for enhanced SERS and FL response. This superstructure supports enhanced aggregation-induced signal amplification for both SERS and FL, offering exceptional sensitivity with LOD as low as 0.0306 pM and 0.409 pM. The efficacy of this method was demonstrated in the evaluation of miRNA-21 in various cancer cell lines.
Assuntos
Técnicas Biossensoriais , DNA , Ouro , Nanopartículas Metálicas , MicroRNAs , Análise Espectral Raman , Humanos , Técnicas Biossensoriais/métodos , MicroRNAs/análise , Ouro/química , Análise Espectral Raman/métodos , Nanopartículas Metálicas/química , DNA/química , Neoplasias , Linhagem Celular Tumoral , Limite de Detecção , Hibridização de Ácido Nucleico , Nanoestruturas/químicaRESUMO
RNA plays an extensive role in a multi-dimensional regulatory system, and its biomedical relationships are scattered across numerous biological studies. However, text mining works dedicated to the extraction of RNA biomedical relations remain limited. In this study, we established a comprehensive and reliable corpus of RNA biomedical relations, recruiting over 30,000 sentences manually curated from more than 15,000 biomedical literature. We also updated RIscoper 2.0, a BERT-based deep learning tool to extract RNA biomedical relation sentences from literature. Benefiting from approximately 100,000 annotated named entities, we integrated the text classification and named entity recognition tasks in this tool. Additionally, RIscoper 2.0 outperformed the original tool in both tasks and can discover new RNA biomedical relations. Additionally, we provided a user-friendly online search tool that enables rapid scanning of RNA biomedical relationships using local and online resources. Both the online tools and data resources of RIscoper 2.0 are available at http://www.rnainter.org/riscoper.