RESUMEN
De novo peptide sequencing is a promising approach for novel peptide discovery, highlighting the performance improvements for the state-of-the-art models. The quality of mass spectra often varies due to unexpected missing of certain ions, presenting a significant challenge in de novo peptide sequencing. Here, we use a novel concept of complementary spectra to enhance ion information of the experimental spectrum and demonstrate it through conceptual and practical analyses. Afterward, we design suitable encoders to encode the experimental spectrum and the corresponding complementary spectrum and propose a de novo sequencing model $\pi$-HelixNovo based on the Transformer architecture. We first demonstrated that $\pi$-HelixNovo outperforms other state-of-the-art models using a series of comparative experiments. Then, we utilized $\pi$-HelixNovo to de novo gut metaproteome peptides for the first time. The results show $\pi$-HelixNovo increases the identification coverage and accuracy of gut metaproteome and enhances the taxonomic resolution of gut metaproteome. We finally trained a powerful $\pi$-HelixNovo utilizing a larger training dataset, and as expected, $\pi$-HelixNovo achieves unprecedented performance, even for peptide-spectrum matches with never-before-seen peptide sequences. We also use the powerful $\pi$-HelixNovo to identify antibody peptides and multi-enzyme cleavage peptides, and $\pi$-HelixNovo is highly robust in these applications. Our results demonstrate the effectivity of the complementary spectrum and take a significant step forward in de novo peptide sequencing.
Asunto(s)
Análisis de Secuencia de Proteína , Espectrometría de Masas en Tándem , Espectrometría de Masas en Tándem/métodos , Análisis de Secuencia de Proteína/métodos , Péptidos , Secuencia de Aminoácidos , Anticuerpos , AlgoritmosRESUMEN
BACKGROUND: MicroRNAs play a critical role in regulating gene expression by binding to specific target sites within gene transcripts, making the identification of microRNA targets a prominent focus of research. Conventional experimental methods for identifying microRNA targets are both time-consuming and expensive, prompting the development of computational tools for target prediction. However, the existing computational tools exhibit limited performance in meeting the demands of practical applications, highlighting the need to improve the performance of microRNA target prediction models. RESULTS: In this paper, we utilize the most popular natural language processing and computer vision technologies to propose a novel approach, called TEC-miTarget, for microRNA target prediction based on transformer encoder and convolutional neural networks. TEC-miTarget treats RNA sequences as a natural language and encodes them using a transformer encoder, a widely used encoder in natural language processing. It then combines the representations of a pair of microRNA and its candidate target site sequences into a contact map, which is a three-dimensional array similar to a multi-channel image. Therefore, the contact map's features are extracted using a four-layer convolutional neural network, enabling the prediction of interactions between microRNA and its candidate target sites. We applied a series of comparative experiments to demonstrate that TEC-miTarget significantly improves microRNA target prediction, compared with existing state-of-the-art models. Our approach is the first approach to perform comparisons with other approaches at both sequence and transcript levels. Furthermore, it is the first approach compared with both deep learning-based and seed-match-based methods. We first compared TEC-miTarget's performance with approaches at the sequence level, and our approach delivers substantial improvements in performance using the same datasets and evaluation metrics. Moreover, we utilized TEC-miTarget to predict microRNA targets in long mRNA sequences, which involves two steps: selecting candidate target site sequences and applying sequence-level predictions. We finally showed that TEC-miTarget outperforms other approaches at the transcript level, including the popular seed match methods widely used in previous years. CONCLUSIONS: We propose a novel approach for predicting microRNA targets at both sequence and transcript levels, and demonstrate that our approach outperforms other methods based on deep learning or seed match. We also provide our approach as an easy-to-use software, TEC-miTarget, at https://github.com/tingpeng17/TEC-miTarget . Our results provide new perspectives for microRNA target prediction.
Asunto(s)
Aprendizaje Profundo , MicroARNs , MicroARNs/genética , MicroARNs/metabolismo , Redes Neurales de la Computación , Programas Informáticos , ARN Mensajero/genéticaRESUMEN
BACKGROUND: In patients with primary biliary cholangitis (PBC) treated with ursodeoxycholic acid (UDCA), the presence of moderate-to-severe interface hepatitis is associated with a higher risk of liver transplantation and death. This highlights the need for novel treatment approaches. In this study, we aimed to investigate whether combination therapy of UDCA and immunosuppressant (IS) was more effective than UDCA monotherapy. METHODS: We conducted a multicenter study involving PBC patients with moderate-to-severe interface hepatitis who underwent paired liver biopsies. Firstly, we compared the efficacy of the combination therapy with UDCA monotherapy on improving biochemistry, histology, survival rates, and prognosis. Subsequently we investigated the predictors of a beneficial response. RESULTS: This retrospective cohort study with prospectively collected data was conducted in China from January 2009 to April 2023. Of the 198 enrolled patients, 32 underwent UDCA monotherapy, while 166 received combination therapy, consisting of UDCA combined with prednisolone, prednisolone plus mycophenolate mofetil (MMF), or prednisolone plus azathioprine (AZA). The monotherapy group was treated for a median duration of 37.6 months (IQR 27.5-58.1), and the combination therapy group had a median treatment duration of 39.3 months (IQR 34.5-48.8). The combination therapy showed a significantly greater efficacy in reducing fibrosis compared to UDCA monotherapy, with an 8.3-fold increase in the regression rate (from 6.3% to 52.4%, P < 0.001). Other parameters, including biochemistry, survival rates, and prognosis, supported its effectiveness. Baseline IgG >1.3 × ULN and ALP <2.4 × ULN were identified as predictors of regression following the combination therapy. A predictive score named FRS, combining these variables, accurately identified individuals achieving fibrosis regression with a cut-off point of ≥ -0.163. The predictive value was validated internally and externally. CONCLUSION: Combination therapy with IS improves outcomes in PBC patients with moderate-to-severe interface hepatitis compared to UDCA monotherapy. Baseline IgG and ALP are the most significant predictors of fibrosis regression. The new predictive score, FRS, incorporating baseline IgG and ALP, can effectively identify individuals who would benefit from the combination therapy.
Asunto(s)
Hepatitis , Cirrosis Hepática Biliar , Humanos , Cirrosis Hepática Biliar/diagnóstico , Cirrosis Hepática Biliar/tratamiento farmacológico , Colagogos y Coleréticos/uso terapéutico , Estudios Retrospectivos , Resultado del Tratamiento , Ácido Ursodesoxicólico/uso terapéutico , Inmunosupresores/uso terapéutico , Prednisolona/uso terapéutico , Terapia de Inmunosupresión , Hepatitis/complicaciones , Inmunoglobulina GRESUMEN
Non-keratinizing carcinoma is the most common subtype of nasopharyngeal carcinoma (NPC). Its poorly differentiated tumor cells and complex microenvironment present challenges to pathological diagnosis. AI-based pathological models have demonstrated potential in diagnosing NPC, but the reliance on costly manual annotation hinders development. To address the challenges, this paper proposes a deep learning-based framework for diagnosing NPC without manual annotation. The framework includes a novel unpaired generative network and a prior-driven image classification system. With pathology-fidelity constraints, the generative network achieves accurate digital staining from H&E to EBER images. The classification system leverages staining specificity and pathological prior knowledge to annotate training data automatically and to classify images for NPC diagnosis. This work used 232 cases for study. The experimental results show that the classification system reached a 99.59% accuracy in classifying EBER images, which closely matched the diagnostic results of pathologists. Utilizing PF-GAN as the backbone of the framework, the system attained a specificity of 0.8826 in generating EBER images, markedly outperforming that of other GANs (0.6137, 0.5815). Furthermore, the F1-Score of the framework for patch level diagnosis was 0.9143, exceeding those of fully supervised models (0.9103, 0.8777). To further validate its clinical efficacy, the framework was compared with experienced pathologists at the WSI level, showing comparable NPC diagnosis performance. This low-cost and precise diagnostic framework optimizes the early pathological diagnosis method for NPC and provides an innovative strategic direction for AI-based cancer diagnosis.
RESUMEN
A label-free optical sandwich immunoassay sensor, utilizing weak value amplification and total internal reflection, was devised for real-time, high-sensitivity analysis and detection of low-concentration targets. 3D printed channels and sodium chloride solution were employed to ensure reproducibility, reliability, and stability of the measurements for calibration. The sandwich structure demonstrated enhanced responsiveness in the proposed optical biosensor through a comparative analysis of the direct assay and sandwich assay for detecting alpha-fetoprotein (AFP) at the same concentration. By optimizing the binding sequences of the coating antibody, target, and detection antibody in the sandwich method, a more suitable sandwich sensing approach based on weak value amplification was achieved. With this approach, the limit of detection (LOD) of 6.29 ng/mL (pM level) for AFP in PBS solution was achieved. AFP testing and regeneration experiments in human serum have proved the feasibility of our methods in detecting complex samples and the reusability of sensing chips. Additionally, the method demonstrated excellent selectivity for unpaired antigens. The efficacy of this methodology was evaluated by simultaneously detecting AFP, carcinoembryonic antigen (CEA), and CA15-3 on a singular sensor chip. In conclusion, the label-free sandwich immunoassay sensing scheme holds promise for advancing the proposed optical sensors based on weak value amplification in early diagnosis and prevention applications. Compared to other biomarker detection methods, it will be easier to promote in practical applications.
Asunto(s)
Técnicas Biosensibles , Antígeno Carcinoembrionario , Límite de Detección , alfa-Fetoproteínas , Técnicas Biosensibles/métodos , alfa-Fetoproteínas/análisis , Humanos , Antígeno Carcinoembrionario/sangre , Antígeno Carcinoembrionario/análisis , Inmunoensayo/métodos , Mucina-1/sangre , Mucina-1/análisis , Anticuerpos Inmovilizados/inmunología , Anticuerpos Inmovilizados/químicaRESUMEN
The public health concern of antibiotic residues in animal-origin food has been a long-standing issue. In this work, we present a novel method for antibiotic detection, leveraging optical weak value amplification and harnessing an indirect competitive inhibition assay, which significantly boosts the system's sensitivity in identifying small molecule antibiotics. We chose chloramphenicol as a model compound and mixed it with chloramphenicol-bovine serum albumin conjugates to bind to the chloramphenicol antibody competitively. We achieved a broad linear detection range of up to 3.24 ng/mL and a high concentration resolution of 33.20 pg/mL. To further validate the universality of our proposed detection methodology, we successfully applied it to testing gibberellin and tetracycline. Moreover, we conducted regeneration experiments and real-sample correlation studies. This study introduces a novel strategy for the label-free optical sensing of small molecule antibiotics, greatly expanding the range of applications for sensors utilizing optical weak value amplification.
RESUMEN
PURPOSE: In pathology images, different stains highlight different glomerular structures, so a supervised deep learning-based glomerular instance segmentation model trained on individual stains performs poorly on other stains. However, it is difficult to obtain a training set with multiple stains because the labeling of pathology images is very time-consuming and tedious. Therefore, in this paper, we proposed an unsupervised stain augmentation-based method for segmentation of glomerular instances. METHODS: In this study, we successfully realized the conversion between different staining methods such as PAS, MT and PASM by contrastive unpaired translation (CUT), thus improving the staining diversity of the training set. Moreover, we replaced the backbone of mask R-CNN with swin transformer to further improve the efficiency of feature extraction and thus achieve better performance in instance segmentation task. RESULTS: To validate the method presented in this paper, we constructed a dataset from 216 WSIs of the three stains in this study. After conducting in-depth experiments, we verified that the instance segmentation method based on stain augmentation outperforms existing methods across all metrics for PAS, PASM, and MT stains. Furthermore, ablation experiments are performed in this paper to further demonstrate the effectiveness of the proposed module. CONCLUSION: This study successfully demonstrated the potential of unsupervised stain augmentation to improve glomerular segmentation in pathology analysis. Future research could extend this approach to other complex segmentation tasks in the pathology image domain to further explore the potential of applying stain augmentation techniques in different domains of pathology image analysis.
RESUMEN
BACKGROUND: Glomerular lesions reflect the onset and progression of renal disease. Pathological diagnoses are widely regarded as the definitive method for recognizing these lesions, as the deviations in histopathological structures closely correlate with impairments in renal function. METHODS: Deep learning plays a crucial role in streamlining the laborious, challenging, and subjective task of recognizing glomerular lesions by pathologists. However, the current methods treat pathology images as data in regular Euclidean space, limiting their ability to efficiently represent the complex local features and global connections. In response to this challenge, this paper proposes a graph neural network (GNN) that utilizes global attention pooling (GAP) to more effectively extract high-level semantic features from glomerular images. The model incorporates Bayesian collaborative learning (BCL), enhancing node feature fine-tuning and fusion during training. In addition, this paper adds a soft classification head to mitigate the semantic ambiguity associated with a purely hard classification. RESULTS: This paper conducted extensive experiments on four glomerular datasets, comprising a total of 491 whole slide images (WSIs) and 9030 images. The results demonstrate that the proposed model achieves impressive F1 scores of 81.37%, 90.12%, 87.72%, and 98.68% on four private datasets for glomerular lesion recognition. These scores surpass the performance of the other models used for comparison. Furthermore, this paper employed a publicly available BReAst Carcinoma Subtyping (BRACS) dataset with an 85.61% F1 score to further prove the superiority of the proposed model. CONCLUSION: The proposed model not only facilitates precise recognition of glomerular lesions but also serves as a potent tool for diagnosing kidney diseases effectively. Furthermore, the framework and training methodology of the GNN can be adeptly applied to address various pathology image classification challenges.
Asunto(s)
Prácticas Interdisciplinarias , Enfermedades Renales , Humanos , Teorema de Bayes , Enfermedades Renales/diagnóstico por imagen , Glomérulos Renales/diagnóstico por imagen , Redes Neurales de la ComputaciónRESUMEN
The demand for accurate and efficient immunoassays calls for the development of precise, high-throughput analysis methods. This paper introduces a novel approach utilizing a weak measurement interface sensor for immunoassays, offering a solution for high throughput analysis. Weak measurement is a precise quantum measurement method that amplifies the weak value of a system in the weak interaction through appropriate pre- and post-selection states. To facilitate the simultaneous analysis of multiple samples, we have developed a chip with six flow channels capable of conducting six immunoassays concurrently. We can perform real-time immunoassay to determine the binding characteristics of spike protein and antibody through real-time analysis of the flow channel images and calculating the relative intensity. The proposed method boasts a simple structure, eliminating the need for intricate nano processes. The spike protein concentration and relative intensity curve were fitted using the Log-Log fitting regression equation, and R2 was 0.91. Utilizing a pre-transformation approach to account for slight variations in detection sensitivity across different flow channels, the present method achieves an impressive limit of detection(LOD) of 0.85 ng/mL for the SARS-CoV-2 the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein, with a system standard deviation of 5.61. Furthermore, this method has been successfully verified for monitoring molecular-specific binding processes and differentiating binding capacities.
Asunto(s)
Técnicas Biosensibles , COVID-19 , SARS-CoV-2 , Glicoproteína de la Espiga del Coronavirus , Glicoproteína de la Espiga del Coronavirus/análisis , Inmunoensayo/métodos , Humanos , COVID-19/diagnóstico , COVID-19/virología , Límite de Detección , Ensayos Analíticos de Alto RendimientoRESUMEN
Allergy is a prevalent disease, and the potential allergic population is expanding with industrialization and changes in people's living standards. Serum immunoglobulin E (IgE) level is one of the critical indicators for determining allergy. Here, we proposed a simple, real-time monitoring, low chip cost, label-free aptamer biosensing strategy based on weak value amplification (WVA) for the quantitative detection of IgE in serum samples, enabling early and accurate diagnosis of allergic or hypersensitive patients. The aptasensor combined an imaging weak measurement system with the high specificity of the aptamer for the marker IgE. By modifying the amino group at the 3-terminal end, the anti-IgE aptamers can attach to a dopamine-modified prism's surface and selectively recognize IgE in human serum. In the presence of IgE, a specific binding reaction occurred, resulting in a change in the refractive index of the reactive region's surface, manifested as a change in the light intensity of the camera acquired experimental images. As the concentration of IgE increased, the relative light intensity advanced sequentially. The WVA-aptasensing strategy achieved a wide detection range of 0.01 ng/mL to 2 µg/mL in phosphate buffered saline buffer, with the resolution as low as 4.3 pg/mL. IgE testing experiments in human serum have proved the feasibility of our methods in detecting complex samples. In addition, the method specifically recognized IgE without interference from other proteins. We believe that our proposed sensing strategy opens up new possibilities for ultrahigh sensitivity screening of IgE and can be expanded to detecting other biomolecules.
Asunto(s)
Aptámeros de Nucleótidos , Técnicas Biosensibles , Inmunoglobulina E , Inmunoglobulina E/sangre , Humanos , Aptámeros de Nucleótidos/química , Técnicas Biosensibles/métodos , Límite de DetecciónRESUMEN
AIMS: The automation of quantitative evaluation for breast immunohistochemistry (IHC) plays a crucial role in reducing the workload of pathologists and enhancing the objectivity of diagnoses. However, current methods face challenges in achieving fully automated immunohistochemistry quantification due to the complexity of segmenting the tumor area into distinct ductal carcinoma in situ (DCIS) and invasive carcinoma (IC) regions. Moreover, the quantitative analysis of immunohistochemistry requires a specific focus on invasive carcinoma regions. METHODS AND RESULTS: In this study, we propose an innovative approach to automatically identify invasive carcinoma regions in breast cancer immunohistochemistry whole-slide images (WSIs). Our method leverages a neural network that combines multi-scale morphological features with boundary features, enabling precise segmentation of invasive carcinoma regions without the need for additional H&E and P63 staining slides. In addition, we introduced an advanced semi-supervised learning algorithm, allowing efficient training of the model using unlabeled data. To evaluate the effectiveness of our approach, we constructed a dataset consisting of 618 IHC-stained WSIs from 170 cases, including four types of staining (ER, PR, HER2, and Ki-67). Notably, the model demonstrated an impressive intersection over union (IoU) score exceeding 80% on the test set. Furthermore, to ascertain the practical utility of our model in IHC quantitative evaluation, we constructed a fully automated Ki-67 scoring system based on the model's predictions. Comparative experiments convincingly demonstrated that our system exhibited high consistency with the scores given by experienced pathologists. CONCLUSIONS: Our developed model excels in accurately distinguishing between DCIS and invasive carcinoma regions in breast cancer immunohistochemistry WSIs. This method paves the way for a clinically available, fully automated immunohistochemistry quantitative scoring system.
RESUMEN
Abstract Objective To investigate the effects of atorvastatin calcium on pulmonary vascular remodeling, the authors explored the regulatory mechanism of Histone Deacetylation Enzyme-2 (HDAC2) in rats with Chronic Obstructive Pulmonary Disease (COPD), and provided a new direction for drug treatment in the progression of vascular remodeling. Methods Eighteen female SD rats were randomly divided into control (Group S1), COPD (Group S2), and atorvastatin calcium + COPD (Group S3) groups. A COPD rat model was established by passive smoking and intratracheal injection of Lipopolysaccharide (LPS). Haematoxylin and eosin staining and Victoria Blue + Van Gibson staining were used to observe pathological changes in the lung tissue. The pulmonary vascular inflammation score was calculated, and the degree of pulmonary vascular remodeling was evaluated. The ratio of Muscular Arteries in lung tissue (MA%), the ratio of the vessel Wall Area to the vessel total area (WA%), and the ratio of the vessel Wall Thickness to the vascular outer diameter (WT%) were measured using imaging software. The expression of HDAC2 was measured using western blotting, ELISA (Enzyme-Linked Immunosorbent Assay), and qPCR (Real-time PCR). Results Compared with the control group, the degree of pulmonary vascular inflammation and pulmonary vascular remodeling increased in rats with COPD. The WT%, WA%, and lung inflammation scores increased significantly; the expression of HDAC2 and HDAC2mRNA in the serum and lung tissue decreased, and the level of Vascular Endothelial Growth Factor (VEGF) in the lung tissues increased (p< 0.05). Compared with the COPD group, the lung tissues from rats in the atorvastatin group had fewer inflammatory cells, and the vascular pathological changes were significantly relieved. The WT%, WA%, and lung inflammation scores decreased significantly; the expression of HDAC2 and HDAC2mRNA in the serum and lung tissues increased, and the level of VEGF in the lung tissues decreased (p< 0.05). Conclusion The present study revealed that atorvastatin calcium could regulate the contents and expression of HDAC2 in serum and lung tissues and inhibit the production of VEGF, thereby regulating pulmonary vascular remodeling in a rat model with COPD.