Multi-scale deep learning for the imbalanced multi-label protein subcellular localization prediction based on immunohistochemistry images

Protein subcellular localization prediction tools

Protein subcellular localization prediction is of great significance in bioinformatics and biological research. Most of the proteins do not have experimentally determined localization information, computational prediction methods and tools have been acting as an active research area for more than two decades now. Knowledge of the subcellular location of a protein provides valuable information about its functionalities, the functioning of the cell, and other possible interactions with proteins. Fast, reliable, and accurate predictors provides platforms to harness the abundance of sequence data to predict subcellular locations accordingly. During the last decade, there has been a considerable amount of research effort aimed at developing subcellular localization predictors. This paper reviews recent subcellular localization prediction tools in the Eukaryotic, Prokaryotic, and Virus-based categories followed by a detailed analysis. Each predictor is discussed based on its main features, strengths, weaknesses, algorithms used, prediction techniques, and analysis. This review is supported by prediction tools taxonomies that highlight their rele- vant area and examples for uncomplicated categorization and ease of understandability. These taxonomies help users find suitable tools according to their needs. Furthermore, recent research gaps and challenges are discussed to cover areas that need the utmost attention. This survey provides an in-depth analysis of the most recent prediction tools to facilitate readers and can be considered a quick guide for researchers to identify and explore the recent literature advancements.

Vislocas: Vision transformers for identifying protein subcellular mis-localization signatures of different cancer subtypes from immunohistochemistry images

Proteins must be sorted to specific subcellular compartments to perform their functions. Abnormal protein subcellular localizations are related to many diseases. Although many efforts have been made in predicting protein subcellular localization from various static information, including sequences, structures and interactions, such static information cannot predict protein mis-localization events in diseases. On the contrary, the IHC (immunohistochemistry) images, which have been widely applied in clinical diagnosis, contains information that can be used to find protein mis-localization events in disease states. In this study, we create the Vislocas method, which is capable of finding mis-localized proteins from IHC images as markers of cancer subtypes. By combining CNNs and vision transformer encoders, Vislocas can automatically extract image features at both global and local level. Vislocas can be trained with full-sized IHC images from scratch. It is the first attempt to create an end-to-end IHC image-based protein subcellular location predictor. Vislocas achieved comparable or better performances than state-of-the-art methods. We applied Vislocas to find significant protein mis-localization events in different subtypes of glioma, melanoma and skin cancer. The mis-localized proteins, which were found purely from IHC images by Vislocas, are in consistency with clinical or experimental results in literatures. All codes of Vislocas have been deposited in a Github repository (https://github.com/JingwenWen99/Vislocas). All datasets of Vislocas have been deposited in Zenodo (https://zenodo.org/records/10632698).

Development and application of a deep learning-based comprehensive early diagnostic model for chronic obstructive pulmonary disease

BACKGROUND

Chronic obstructive pulmonary disease (COPD) is a frequently diagnosed yet treatable condition, provided it is identified early and managed effectively. This study aims to develop an advanced COPD diagnostic model by integrating deep learning and radiomics features.

METHODS

We utilized a dataset comprising CT images from 2,983 participants, of which 2,317 participants also provided epidemiological data through questionnaires. Deep learning features were extracted using a Variational Autoencoder, and radiomics features were obtained using the PyRadiomics package. Multi-Layer Perceptrons were used to construct models based on deep learning and radiomics features independently, as well as a fusion model integrating both. Subsequently, epidemiological questionnaire data were incorporated to establish a more comprehensive model. The diagnostic performance of standalone models, the fusion model and the comprehensive model was evaluated and compared using metrics including accuracy, precision, recall, F1-score, Brier score, receiver operating characteristic curves, and area under the curve (AUC).

RESULTS

The fusion model exhibited outstanding performance with an AUC of 0.952, surpassing the standalone models based solely on deep learning features (AUC = 0.844) or radiomics features (AUC = 0.944). Notably, the comprehensive model, incorporating deep learning features, radiomics features, and questionnaire variables demonstrated the highest diagnostic performance among all models, yielding an AUC of 0.971.

CONCLUSION

We developed and implemented a data fusion strategy to construct a state-of-the-art COPD diagnostic model integrating deep learning features, radiomics features, and questionnaire variables. Our data fusion strategy proved effective, and the model can be easily deployed in clinical settings.

TRIAL REGISTRATION

Not applicable. This study is NOT a clinical trial, it does not report the results of a health care intervention on human participants.

Automated identification of protein expression intensity and classification of protein cellular locations in mouse brain regions from immunofluorescence images

Knowledge of protein expression in mammalian brains at regional and cellular levels can facilitate understanding of protein functions and associated diseases. As the mouse brain is a typical mammalian brain considering cell type and structure, several studies have been conducted to analyze protein expression in mouse brains. However, labeling protein expression using biotechnology is costly and time-consuming. Therefore, automated models that can accurately recognize protein expression are needed. Here, we constructed machine learning models to automatically annotate the protein expression intensity and cellular location in different mouse brain regions from immunofluorescence images. The brain regions and sub-regions were segmented through learning image features using an autoencoder and then performing K-means clustering and registration to align with the anatomical references. The protein expression intensities for those segmented structures were computed on the basis of the statistics of the image pixels, and patch-based weakly supervised methods and multi-instance learning were used to classify the cellular locations. Results demonstrated that the models achieved high accuracy in the expression intensity estimation, and the F1 score of the cellular location prediction was 74.5%. This work established an automated pipeline for analyzing mouse brain images and provided a foundation for further study of protein expression and functions.

A Review for Artificial Intelligence Based Protein Subcellular Localization

Proteins need to be located in appropriate spatiotemporal contexts to carry out their diverse biological functions. Mislocalized proteins may lead to a broad range of diseases, such as cancer and Alzheimer's disease. Knowing where a target protein resides within a cell will give insights into tailored drug design for a disease. As the gold validation standard, the conventional wet lab uses fluorescent microscopy imaging, immunoelectron microscopy, and fluorescent biomarker tags for protein subcellular location identification. However, the booming era of proteomics and high-throughput sequencing generates tons of newly discovered proteins, making protein subcellular localization by wet-lab experiments a mission impossible. To tackle this concern, in the past decades, artificial intelligence (AI) and machine learning (ML), especially deep learning methods, have made significant progress in this research area. In this article, we review the latest advances in AI-based method development in three typical types of approaches, including sequence-based, knowledge-based, and image-based methods. We also elaborately discuss existing challenges and future directions in AI-based method development in this research field.

Dual-Signal Feature Spaces Map Protein Subcellular Locations Based on Immunohistochemistry Image and Protein Sequence

Protein is one of the primary biochemical macromolecular regulators in the compartmental cellular structure, and the subcellular locations of proteins can therefore provide information on the function of subcellular structures and physiological environments. Recently, data-driven systems have been developed to predict the subcellular location of proteins based on protein sequence, immunohistochemistry (IHC) images, or immunofluorescence (IF) images. However, the research on the fusion of multiple protein signals has received little attention. In this study, we developed a dual-signal computational protocol by incorporating IHC images into protein sequences to learn protein subcellular localization. Three major steps can be summarized as follows in this protocol: first, a benchmark database that includes 281 proteins sorted out from 4722 proteins of the Human Protein Atlas (HPA) and Swiss-Prot database, which is involved in the endoplasmic reticulum (ER), Golgi apparatus, cytosol, and nucleoplasm; second, discriminative feature operators were first employed to quantitate protein image-sequence samples that include IHC images and protein sequence; finally, the feature subspace of different protein signals is absorbed to construct multiple sub-classifiers via dimensionality reduction and binary relevance (BR), and multiple confidence derived from multiple sub-classifiers is adopted to decide subcellular location by the centralized voting mechanism at the decision layer. The experimental results indicated that the dual-signal model embedded IHC images and protein sequences outperformed the single-signal models with accuracy, precision, and recall of 75.41%, 80.38%, and 74.38%, respectively. It is enlightening for further research on protein subcellular location prediction under multi-signal fusion of protein.

RDDL: A systematic ensemble pipeline tool that streamlines balancing training schemes to reduce the effects of data imbalance in rare-disease-related deep-learning applications

Identifying lowly prevalent diseases, or rare diseases, in their early stages is key to disease treatment in the medical field. Deep learning techniques now provide promising tools for this purpose. Nevertheless, the low prevalence of rare diseases entangles the proper application of deep networks for disease identification due to the severe class-imbalance issue. In the past decades, some balancing methods have been studied to handle the data-imbalance issue. The bad news is that it is verified that none of these methods guarantees superior performance to others. This performance variation causes the need to formulate a systematic pipeline with a comprehensive software tool for enhancing deep-learning applications in rare disease identification. We reviewed the existing balancing schemes and summarized a systematic deep ensemble pipeline with a constructed tool called RDDL for handling the data imbalance issue. Through two real case studies, we showed that rare disease identification could be boosted with this systematic RDDL pipeline tool by lessening the data imbalance problem during model training. The RDDL pipeline tool is available at https://github.com/cobisLab/RDDL/.

Different Recognition of Protein Features Depending on Deep Learning Models: A Case Study of Aromatic Decarboxylase UbiD

The number of unannotated protein sequences is explosively increasing due to genome sequence technology. A more comprehensive understanding of protein functions for protein annotation requires the discovery of new features that cannot be captured from conventional methods. Deep learning can extract important features from input data and predict protein functions based on the features. Here, protein feature vectors generated by 3 deep learning models are analyzed using Integrated Gradients to explore important features of amino acid sites. As a case study, prediction and feature extraction models for UbiD enzymes were built using these models. The important amino acid residues extracted from the models were different from secondary structures, conserved regions and active sites of known UbiD information. Interestingly, the different amino acid residues within UbiD sequences were regarded as important factors depending on the type of models and sequences. The Transformer models focused on more specific regions than the other models. These results suggest that each deep learning model understands protein features with different aspects from existing knowledge and has the potential to discover new laws of protein functions. This study will help to extract new protein features for the other protein annotations.

Computational prediction of protein folding rate using structural parameters and network centrality measures

Protein folding is a complex physicochemical process whereby a polymer of amino acids samples numerous conformations in its unfolded state before settling on an essentially unique native three-dimensional (3D) structure. To understand this process, several theoretical studies have used a set of 3D structures, identified different structural parameters, and analyzed their relationships using the natural logarithmic protein folding rate (ln(k_f)). Unfortunately, these structural parameters are specific to a small set of proteins that are not capable of accurately predicting ln(k_f) for both two-state (TS) and non-two-state (NTS) proteins. To overcome the limitations of the statistical approach, a few machine learning (ML)-based models have been proposed using limited training data. However, none of these methods can explain plausible folding mechanisms. In this study, we evaluated the predictive capabilities of ten different ML algorithms using eight different structural parameters and five different network centrality measures based on newly constructed datasets. In comparison to the other nine regressors, support vector machine was found to be the most appropriate for predicting ln(k_f) with mean absolute differences of 1.856, 1.55, and 1.745 for the TS, NTS, and combined datasets, respectively. Furthermore, combining structural parameters and network centrality measures improves the prediction performance compared to individual parameters, indicating that multiple factors are involved in the folding process.

SiameseCPP: a sequence-based Siamese network to predict cell-penetrating peptides by contrastive learning

BACKGROUND

Cell-penetrating peptides (CPPs) have received considerable attention as a means of transporting pharmacologically active molecules into living cells without damaging the cell membrane, and thus hold great promise as future therapeutics. Recently, several machine learning-based algorithms have been proposed for predicting CPPs. However, most existing predictive methods do not consider the agreement (disagreement) between similar (dissimilar) CPPs and depend heavily on expert knowledge-based handcrafted features.

RESULTS

In this study, we present SiameseCPP, a novel deep learning framework for automated CPPs prediction. SiameseCPP learns discriminative representations of CPPs based on a well-pretrained model and a Siamese neural network consisting of a transformer and gated recurrent units. Contrastive learning is used for the first time to build a CPP predictive model. Comprehensive experiments demonstrate that our proposed SiameseCPP is superior to existing baseline models for predicting CPPs. Moreover, SiameseCPP also achieves good performance on other functional peptide datasets, exhibiting satisfactory generalization ability.

SPECHT: Self-tuning Plausibility based object detection Enables quantification of Conflict in Heterogeneous multi-scale microscopy

Identification of small objects in fluorescence microscopy is a non-trivial task burdened by parameter-sensitive algorithms, for which there is a clear need for an approach that adapts dynamically to changing imaging conditions. Here, we introduce an adaptive object detection method that, given a microscopy image and an image level label, uses kurtosis-based matching of the distribution of the image differential to express operator intent in terms of recall or precision. We show how a theoretical upper bound of the statistical distance in feature space enables application of belief theory to obtain statistical support for each detected object, capturing those aspects of the image that support the label, and to what extent. We validate our method on 2 datasets: distinguishing sub-diffraction limit caveolae and scaffold by stimulated emission depletion (STED) super-resolution microscopy; and detecting amyloid-β deposits in confocal microscopy retinal cross-sections of neuropathologically confirmed Alzheimer's disease donor tissue. Our results are consistent with biological ground truth and with previous subcellular object classification results, and add insight into more nuanced class transition dynamics. We illustrate the novel application of belief theory to object detection in heterogeneous microscopy datasets and the quantification of conflict of evidence in a joint belief function. By applying our method successfully to diffraction-limited confocal imaging of tissue sections and super-resolution microscopy of subcellular structures, we demonstrate multi-scale applicability.

Multiple Parallel Fusion Network for Predicting Protein Subcellular Localization from Stimulated Raman Scattering (SRS) Microscopy Images in Living Cells

Stimulated Raman Scattering Microscopy (SRS) is a powerful tool for label-free detailed recognition and investigation of the cellular and subcellular structures of living cells. Determining subcellular protein localization from the cell level of SRS images is one of the basic goals of cell biology, which can not only provide useful clues for their functions and biological processes but also help to determine the priority and select the appropriate target for drug development. However, the bottleneck in predicting subcellular protein locations of SRS cell imaging lies in modeling complicated relationships concealed beneath the original cell imaging data owing to the spectral overlap information from different protein molecules. In this work, a multiple parallel fusion network, MPFnetwork, is proposed to study the subcellular locations from SRS images. This model used a multiple parallel fusion model to construct feature representations and combined multiple nonlinear decomposing algorithms as the automated subcellular detection method. Our experimental results showed that the MPFnetwork could achieve over 0.93 dice correlation between estimated and true fractions on SRS lung cancer cell datasets. In addition, we applied the MPFnetwork method to cell images for label-free prediction of several different subcellular components simultaneously, rather than using several fluorescent labels. These results open up a new method for the time-resolved study of subcellular components in different cells, especially cancer cells.

GraphLoc: a graph neural network model for predicting protein subcellular localization from immunohistochemistry images

Motivation

Recognition of protein subcellular distribution patterns and identification of location biomarker proteins in cancer tissues are important for understanding protein functions and related diseases. Immunohistochemical (IHC) images enable visualizing the distribution of proteins at the tissue level, providing an important resource for the protein localization studies. In the past decades, several image-based protein subcellular location prediction methods have been developed, but the prediction accuracies still have much space to improve due to the complexity of protein patterns resulting from multi-label proteins and variation of location patterns across cell types or states.

Results

Here, we propose a multi-label multi-instance model based on deep graph convolutional neural networks, GraphLoc, to recognize protein subcellular location patterns. GraphLoc builds a graph of multiple IHC images for one protein, learns protein-level representations by graph convolutions, and predicts multi-label information by a dynamic threshold method. Our results show that GraphLoc is a promising model for image-based protein subcellular location prediction with model interpretability. Furthermore, we apply GraphLoc to the identification of candidate location biomarkers and potential members for protein networks. A large portion of the predicted results have supporting evidence from the existing literatures and the new candidates also provide guidance for further experimental screening.

Availability

The dataset and code are available at: www.csbio.sjtu.edu.cn/bioinf/GraphLoc.

Supplementary information

Supplementary data are available at Bioinformatics online.

Recent Advances in the Prediction of Subcellular Localization of Proteins and Related Topics

Prediction of subcellular localization of proteins from their amino acid sequences has a long history in bioinformatics and is still actively developing, incorporating the latest advances in machine learning and proteomics. Notably, deep learning-based methods for natural language processing have made great contributions. Here, we review recent advances in the field as well as its related fields, such as subcellular proteomics and the prediction/recognition of subcellular localization from image data.