Epitome: predicting epigenetic events in novel cell types with multi-cell deep ensemble learning.
Ontology highlight
ABSTRACT:
SUBMITTER: Morrow AK
PROVIDER: S-EPMC8565335 | biostudies-literature | 2021 Nov
REPOSITORIES: biostudies-literature
Publications
Epitome: predicting epigenetic events in novel cell types with multi-cell deep ensemble learning.
Morrow Alyssa Kramer AK Hughes John Weston JW Singh Jahnavi J Joseph Anthony Douglas AD Yosef Nir N
Nucleic acids research 20211101 19
The accumulation of large epigenomics data consortiums provides us with the opportunity to extrapolate existing knowledge to new cell types and conditions. We propose Epitome, a deep neural network that learns similarities of chromatin accessibility between well characterized reference cell types and a query cellular context, and copies over signal of transcription factor binding and modification of histones from reference cell types when chromatin profiles are similar to the query. Epitome achi ...[more]
Similar Datasets
Project description:DNNs trained for predicting cellular events from DNA sequence have become emerging tools to help elucidate biological mechanisms underlying associations identified in genome-wide association studies. To enhance the training, multi-task learning (MTL) has been commonly exploited in previous works where trained networks were needed for multiple profiles differing in either event modality or cell type. All existing works adopted a simple MTL framework where all tasks share a single feature extraction network. Such a strategy even though effective to a certain extent leads to substantial negative transfer, meaning the existence of a large portion of tasks for which models obtained through MTL perform worse than those by single-task learning. There have been methods developed to address such negative transfer in other domains, such as computer vision. However, these methods are generally with limited scalability. In this paper, we propose a highly scalable task grouping framework to address negative transfer by only jointly training tasks that are potentially beneficial to each other. The proposed method exploits the network weights associated with task-specific classification heads that can be cheaply obtained by one-time joint training of all tasks. Our results using a dataset consisting of 367 epigenetic profiles demonstrate the effectiveness of the proposed approach and its superiority over baseline methods.
| S-EPMC10779439 | biostudies-literature
Project description:BackgroundPredicting drug side effects is an important topic in the drug discovery. Although several machine learning methods have been proposed to predict side effects, there is still space for improvements. Firstly, the side effect prediction is a multi-label learning task, and we can adopt the multi-label learning techniques for it. Secondly, drug-related features are associated with side effects, and feature dimensions have specific biological meanings. Recognizing critical dimensions and reducing irrelevant dimensions may help to reveal the causes of side effects.MethodsIn this paper, we propose a novel method 'feature selection-based multi-label k-nearest neighbor method' (FS-MLKNN), which can simultaneously determine critical feature dimensions and construct high-accuracy multi-label prediction models.ResultsComputational experiments demonstrate that FS-MLKNN leads to good performances as well as explainable results. To achieve better performances, we further develop the ensemble learning model by integrating individual feature-based FS-MLKNN models. When compared with other state-of-the-art methods, the ensemble method produces better performances on benchmark datasets.ConclusionsIn conclusion, FS-MLKNN and the ensemble method are promising tools for the side effect prediction. The source code and datasets are available in the Additional file 1.
| S-EPMC4634905 | biostudies-literature
Project description:Recent studies have identified G protein-coupled receptor 40 (GPR40) as a promising target for treating type 2 diabetes mellitus, and GPR40 agonists have several superior effects over other hypoglycemic drugs, including cardiovascular protection and suppression of glucagon levels. In this study, we constructed an up-to-date GPR40 ligand dataset for training models and performed a systematic optimization of the ensemble model, resulting in a powerful ensemble model (ROC AUC: 0.9496) for distinguishing GPR40 agonists and non-agonists. The ensemble model is divided into three layers, and the optimization process is carried out in each layer. We believe that these results will prove helpful for both the development of GPR40 agonists and ensemble models. All the data and models are available on GitHub. (https://github.com/Jiamin-Yang/ensemble_model).
| S-EPMC10661831 | biostudies-literature
Project description:BackgroundCervical cell segmentation is a fundamental step in automated cervical cancer cytology screening. The aim of this study was to develop and evaluate a deep ensemble model for cervical cell segmentation including both cytoplasm and nucleus segmentation.MethodsThe Cx22 dataset was used to develop the automated cervical cell segmentation algorithm. The U-Net, U-Net + + , DeepLabV3, DeepLabV3Plus, Transunet, and Segformer were used as candidate model architectures, and each of the first four architectures adopted two different encoders choosing from resnet34, resnet50 and denseNet121. Models were trained under two settings: trained from scratch, encoders initialized from ImageNet pre-trained models and then all layers were fine-tuned. For every segmentation task, four models were chosen as base models, and Unweighted average was adopted as the model ensemble method.ResultsU-Net and U-Net + + with resnet34 and denseNet121 encoders trained using transfer learning consistently performed better than other models, so they were chosen as base models. The ensemble model obtained the Dice similarity coefficient, sensitivity, specificity of 0.9535 (95% CI:0.9534-0.9536), 0.9621 (0.9619-0.9622),0.9835 (0.9834-0.9836) and 0.7863 (0.7851-0.7876), 0.9581 (0.9573-0.959), 0.9961 (0.9961-0.9962) on cytoplasm segmentation and nucleus segmentation, respectively. The Dice, sensitivity, specificity of baseline models for cytoplasm segmentation and nucleus segmentation were 0.948, 0.954, 0.9823 and 0.750, 0.713, 0.9988, respectively. Except for the specificity of cytoplasm segmentation, all metrics outperformed the best baseline models (P < 0.05) with a moderate margin.ConclusionsThe proposed algorithm achieved better performances on cervical cell segmentation than baseline models. It can be potentially used in automated cervical cancer cytology screening system.
| S-EPMC10514950 | biostudies-literature
Project description:N6-methyladenosine (m6A) is the most abundant within eukaryotic messenger RNA modification, which plays an essential regulatory role in the control of cellular functions and gene expression. However, it remains an outstanding challenge to detect mRNA m6A transcriptome-wide at base resolution via experimental approaches, which are generally time-consuming and expensive. Developing computational methods is a good strategy for accurate in silico detection of m6A modification sites from the large amount of RNA sequence data. Unfortunately, the existing computational models are usually only for m6A site prediction in a single species, without considering the tissue level of species, while most of them are constructed based on low-confidence level data generated by an m6A antibody immunoprecipitation (IP)-based sequencing method, thereby restricting reliability and generalizability of proposed models. Here, we review recent advances in computational prediction of m6A sites and construct a new computational approach named im6APred using ensemble deep learning to accurately identify m6A sites based on high-confidence level data in multiple tissues of mammals. Our model im6APred builds upon a comprehensive evaluation of multiple classification methods, including four traditional classification algorithms and three deep learning methods and their ensembles. The optimal base-classifier combinations are then chosen by five-fold cross-validation test to achieve an effective stacked model. Our model im6APred can produce the area under the receiver operating characteristic curve (AUROC) in the range of 0.82-0.91 on independent tests, indicating that our model has the ability to learn general methylation rules on RNA bases and generalize to m6A transcriptome-wide identification. Moreover, AUROCs in the range of 0.77-0.96 were achieved using cross-species/tissues validation on the benchmark dataset, demonstrating differences in predictive performance at the tissue level and the need for constructing tissue-specific models for m6A site prediction.
| S-EPMC9778682 | biostudies-literature
Project description:IntroductionDespite the many benefits immunotherapy has brought to patients with different cancers, its clinical applications and improvements are still hindered by drug resistance. Fostering a reliable approach to identifying sufferers who are sensitive to certain immunotherapeutic agents is of great clinical relevance.MethodsWe propose an ELISE (Ensemble Learning for Immunotherapeutic Response Evaluation) pipeline to generate a robust and highly accurate approach to predicting individual responses to immunotherapies. ELISE employed iterative univariable logistic regression to select genetic features of patients, using Monte Carlo Tree Search (MCTS) to tune hyperparameters. In each trial, ELISE selected multiple models for integration based on add or concatenate stacking strategies, including deep neural network, automatic feature interaction learning via self-attentive neural networks, deep factorization machine, compressed interaction network, and linear neural network, then adopted the best trial to generate a final approach. SHapley Additive exPlanations (SHAP) algorithm was applied to interpret ELISE, which was then validated in an independent test set.ResultRegarding prediction of responses to atezolizumab within esophageal adenocarcinoma (EAC) patients, ELISE demonstrated a superior accuracy (Area Under Curve [AUC] = 100.00%). AC005786.3 (Mean [|SHAP value|] = 0.0097) was distinguished as the most valuable contributor to ELISE output, followed by SNORD3D (0.0092), RN7SKP72 (0.0081), EREG (0.0069), IGHV4-80 (0.0063), and MIR4526 (0.0063). Mechanistically, immunoglobulin complex, immunoglobulin production, adaptive immune response, antigen binding and others, were downregulated in ELISE-neg EAC subtypes and resulted in unfavorable responses. More encouragingly, ELISE could be extended to accurately estimate the responsiveness of various immunotherapeutic agents against other cancers, including PD1/PD-L1 suppressor against metastatic urothelial cancer (AUC = 88.86%), and MAGE-A3 immunotherapy against metastatic melanoma (AUC = 100.00%).DiscussionThis study presented deep insights into integrating ensemble deep learning with self-attention as a mechanism for predicting immunotherapy responses to human cancers, highlighting ELISE as a potential tool to generate reliable approaches to individualized treatment.
| S-EPMC9751999 | biostudies-literature
Project description:Detecting cancers at early stages can dramatically reduce mortality rates. Therefore, practical cancer screening at the population level is needed. To develop a comprehensive detection system to classify multiple cancer types, we integrated an artificial intelligence deep learning neural network and noncoding RNA biomarkers selected from massive data. Our system can accurately detect cancer vs. healthy objects with 96.3% of AUC of ROC (Area Under Curve of a Receiver Operating Characteristic curve), and it surprisingly reaches 78.77% of AUC when validated by real-world raw data from a completely independent data set. Even validating with raw exosome data from blood, our system can reach 72% of AUC. Moreover, our system significantly outperforms conventional machine learning models, such as random forest. Intriguingly, with no more than six biomarkers, our approach can easily discriminate any individual cancer type vs. normal with 99% to 100% AUC. Furthermore, a comprehensive marker panel can simultaneously multi-classify common cancers with a stable 82.15% accuracy rate for heterogeneous cancerous tissues and conditions. This detection system provides a promising practical framework for automatic cancer screening at population level. Key points: (1) We developed a practical cancer screening system, which is simple, accurate, affordable, and easy to operate. (2) Our system binarily classify cancers vs. normal with >96% AUC. (3) In total, 26 individual cancer types can be easily detected by our system with 99 to 100% AUC. (4) The system can detect multiple cancer types simultaneously with >82% accuracy.
| S-EPMC8774129 | biostudies-literature
Project description:Single-cell RNA-sequencing (scRNA-seq) is a powerful technique that provides high-resolution expression profiling of individual cells. It significantly advances our understanding of cellular diversity and function. Despite its potential, the analysis of scRNA-seq data poses considerable challenges related to multicollinearity, data imbalance, and batch effect. One of the pivotal tasks in single-cell data analysis is cell type annotation, which classifies cells into discrete types based on their gene expression profiles. In this work, we propose a novel modeling formalism for cell type annotation with a supervised contrastive learning method, named SCLSC (Supervised Contrastive Learning for Single Cell). Different from the previous usage of contrastive learning in single cell data analysis, we employed the contrastive learning for instance-type pairs instead of instance-instance pairs. More specifically, in the cell type annotation task, the contrastive learning is applied to learn cell and cell type representation that render cells of the same type to be clustered in the new embedding space. Through this approach, the knowledge derived from annotated cells is transferred to the feature representation for scRNA-seq data. The whole training process becomes more efficient when conducting contrastive learning for cell and their types. Our experiment results demonstrate that the proposed SCLSC method consistently achieves superior accuracy in predicting cell types compared to five state-of-the-art methods. SCLSC also performs well in identifying cell types in different batch groups. The simplicity of our method allows for scalability, making it suitable for analyzing datasets with a large number of cells. In a real-world application of SCLSC to monitor the dynamics of immune cell subpopulations over time, SCLSC demonstrates a capability to discriminate cell subtypes of CD19+ B cells that were not present in the training dataset.
| S-EPMC10764802 | biostudies-literature
Project description:MotivationSingle-cell clustering plays a crucial role in distinguishing between cell types, facilitating the analysis of cell heterogeneity mechanisms. While many existing clustering methods rely solely on gene expression data obtained from single-cell RNA sequencing techniques to identify cell clusters, the information contained in mono-omic data is often limited, leading to suboptimal clustering performance. The emergence of single-cell multi-omics sequencing technologies enables the integration of multiple omics data for identifying cell clusters, but how to integrate different omics data effectively remains challenging. In addition, designing a clustering method that performs well across various types of multi-omics data poses a persistent challenge due to the data's inherent characteristics.ResultsIn this paper, we propose a graph-regularized multi-view ensemble clustering (GRMEC-SC) model for single-cell clustering. Our proposed approach can adaptively integrate multiple omics data and leverage insights from multiple base clustering results. We extensively evaluate our method on five multi-omics datasets through a series of rigorous experiments. The results of these experiments demonstrate that our GRMEC-SC model achieves competitive performance across diverse multi-omics datasets with varying characteristics.Availability and implementationImplementation of GRMEC-SC, along with examples, can be found on the GitHub repository: https://github.com/polarisChen/GRMEC-SC.
| S-EPMC11015955 | biostudies-literature
Project description:Multi-omics data are good resources for prognosis and survival prediction; however, these are difficult to integrate computationally. We introduce DeepProg, a novel ensemble framework of deep-learning and machine-learning approaches that robustly predicts patient survival subtypes using multi-omics data. It identifies two optimal survival subtypes in most cancers and yields significantly better risk-stratification than other multi-omics integration methods. DeepProg is highly predictive, exemplified by two liver cancer (C-index 0.73-0.80) and five breast cancer datasets (C-index 0.68-0.73). Pan-cancer analysis associates common genomic signatures in poor survival subtypes with extracellular matrix modeling, immune deregulation, and mitosis processes. DeepProg is freely available at https://github.com/lanagarmire/DeepProg.
| S-EPMC8281595 | biostudies-literature