DTL-DephosSite: Deep Transfer Learning Based Approach to Predict Dephosphorylation Sites.
Ontology highlight
ABSTRACT:
SUBMITTER: Chaudhari M
PROVIDER: S-EPMC8264445 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Protein dephosphorylation, which is an inverse process of phosphorylation, plays a crucial role in a myriad of cellular processes, including mitotic cycle, proliferation, differentiation, and cell growth. Compared with tyrosine kinase substrate and phosphorylation site prediction, there is a paucity of studies focusing on computational methods of predicting protein tyrosine phosphatase substrates and dephosphorylation sites. In this work, we developed two elegant models for predicting the substrate dephosphorylation sites of three specific phosphatases, namely, PTP1B, SHP-1, and SHP-2. The first predictor is called MGPS-DEPHOS, which is modified from the GPS (Group-based Prediction System) algorithm with an interpretable capability. The second predictor is called CKSAAP-DEPHOS, which is built through the combination of support vector machine (SVM) and the composition of k-spaced amino acid pairs (CKSAAP) encoding scheme. Benchmarking experiments using jackknife cross validation and 30 repeats of 5-fold cross validation tests show that MGPS-DEPHOS and CKSAAP-DEPHOS achieved AUC values of 0.921, 0.914 and 0.912, for predicting dephosphorylation sites of the three phosphatases PTP1B, SHP-1, and SHP-2, respectively. Both methods outperformed the previously developed kNN-DEPHOS algorithm. In addition, a web server implementing our algorithms is publicly available at http://genomics.fzu.edu.cn/dephossite/ for the research community.
| S-EPMC4802303 | biostudies-literature
Project description:DNA N4-methylcytosine (4mC) is an important epigenetic modification that plays a vital role in regulating DNA replication and expression. However, it is challenging to detect 4mC sites through experimental methods, which are time-consuming and costly. Thus, computational tools that can identify 4mC sites would be very useful for understanding the mechanism of this important type of DNA modification. Several machine learning-based 4mC predictors have been proposed in the past 3 years, although their performance is unsatisfactory. Deep learning is a promising technique for the development of more accurate 4mC site predictions. In this work, we propose a deep learning-based approach, called DeepTorrent, for improved prediction of 4mC sites from DNA sequences. It combines four different feature encoding schemes to encode raw DNA sequences and employs multi-layer convolutional neural networks with an inception module integrated with bidirectional long short-term memory to effectively learn the higher-order feature representations. Dimension reduction and concatenated feature maps from the filters of different sizes are then applied to the inception module. In addition, an attention mechanism and transfer learning techniques are also employed to train the robust predictor. Extensive benchmarking experiments demonstrate that DeepTorrent significantly improves the performance of 4mC site prediction compared with several state-of-the-art methods.
| S-EPMC8599298 | biostudies-literature
Project description:Protein phosphorylation, which is one of the most important post-translational modifications (PTMs), is involved in regulating myriad cellular processes. Herein, we present a novel deep learning based approach for organism-specific protein phosphorylation site prediction in Chlamydomonas reinhardtii, a model algal phototroph. An ensemble model combining convolutional neural networks and long short-term memory (LSTM) achieves the best performance in predicting phosphorylation sites in C. reinhardtii. Deemed Chlamy-EnPhosSite, the measured best AUC and MCC are 0.90 and 0.64 respectively for a combined dataset of serine (S) and threonine (T) in independent testing higher than those measures for other predictors. When applied to the entire C. reinhardtii proteome (totaling 1,809,304 S and T sites), Chlamy-EnPhosSite yielded 499,411 phosphorylated sites with a cut-off value of 0.5 and 237,949 phosphorylated sites with a cut-off value of 0.7. These predictions were compared to an experimental dataset of phosphosites identified by liquid chromatography-tandem mass spectrometry (LC-MS/MS) in a blinded study and approximately 89.69% of 2,663 C. reinhardtii S and T phosphorylation sites were successfully predicted by Chlamy-EnPhosSite at a probability cut-off of 0.5 and 76.83% of sites were successfully identified at a more stringent 0.7 cut-off. Interestingly, Chlamy-EnPhosSite also successfully predicted experimentally confirmed phosphorylation sites in a protein sequence (e.g., RPS6 S245) which did not appear in the training dataset, highlighting prediction accuracy and the power of leveraging predictions to identify biologically relevant PTM sites. These results demonstrate that our method represents a robust and complementary technique for high-throughput phosphorylation site prediction in C. reinhardtii. It has potential to serve as a useful tool to the community. Chlamy-EnPhosSite will contribute to the understanding of how protein phosphorylation influences various biological processes in this important model microalga.
| S-EPMC8206365 | biostudies-literature
Project description:Snake venom contains many toxic proteins that can destroy the circulatory system or nervous system of prey. Studies have found that these snake venom proteins have the potential to treat cardiovascular and nervous system diseases. Therefore, the study of snake venom protein is conducive to the development of related drugs. The research technologies based on traditional biochemistry can accurately identify these proteins, but the experimental cost is high and the time is long. Artificial intelligence technology provides a new means and strategy for large-scale screening of snake venom proteins from the perspective of computing. In this paper, we developed a sequence-based computational method to recognize snake toxin proteins. Specially, we utilized three different feature descriptors, namely g-gap, natural vector and word 2 vector, to encode snake toxin protein sequences. The analysis of variance (ANOVA), gradient-boost decision tree algorithm (GBDT) combined with incremental feature selection (IFS) were used to optimize the features, and then the optimized features were input into the deep learning model for model training. The results show that our model can achieve a prediction performance with an accuracy of 82.00% in 10-fold cross-validation. The model is further verified on independent data, and the accuracy rate reaches to 81.14%, which demonstrated that our model has excellent prediction performance and robustness.
| S-EPMC10829051 | biostudies-literature
Project description:Methylation, which is one of the most prominent post-translational modifications on proteins, regulates many important cellular functions. Though several model-based methylation site predictors have been reported, all existing methods employ machine learning strategies, such as support vector machines and random forest, to predict sites of methylation based on a set of "hand-selected" features. As a consequence, the subsequent models may be biased toward one set of features. Moreover, due to the large number of features, model development can often be computationally expensive. In this paper, we propose an alternative approach based on deep learning to predict arginine methylation sites. Our model, which we termed DeepRMethylSite, is computationally less expensive than traditional feature-based methods while eliminating potential biases that can arise through features selection. Based on independent testing on our dataset, DeepRMethylSite achieved efficiency scores of 68%, 82% and 0.51 with respect to sensitivity (SN), specificity (SP) and Matthew's correlation coefficient (MCC), respectively. Importantly, in side-by-side comparisons with other state-of-the-art methylation site predictors, our method performs on par or better in all scoring metrics tested.
| S-EPMC7554107 | biostudies-literature
Project description:Due to this epidemic of COVID-19, the everyday lives, welfare, and wealth of a country are affected. Inefficiency, a lack of medical diagnostics, and inadequately trained healthcare professionals are among the most significant barriers to arresting the development of this disease. Blockchain offers enormous promise for providing consistent and reliable real time and smart health facilities offsite. The infected patients with COVID-19 have shown they often have a lung infection upon arrival. It can be detected and analyzed using CT scan images. Unfortunately, though, it is time-consuming and liable to error. Thus, the assessment of chest CT scans must be automated. The proposed method uses transfer deep learning techniques to analyze CT scan images automatically. Transfer deep learning can improve the parameters of networks on huge databases, and pretrained networks can be used effectively on small datasets. We proposed a model built on VGGNet19, a convolutional neural network to classify individuals infected with coronavirus utilizing images of CT radiographs. We have used a globally accessible CT scan database that included 2500 CT pictures with COVID-19 infection and 2500 CT images without COVID-19 infection. An extensive experiment has been conducted using three deep learning methods such as VGG19, Xception Net, and CNN. Experiment findings indicate that the proposed model outperforms the other Xception Net and CNN models considerably. The results demonstrate that the proposed models have an accuracy of up to 95% and area under the receiver operating characteristic curve up to 95%.
| S-EPMC9212254 | biostudies-literature
Project description:MicroRNAs (miRNAs) are small non-coding RNAs that regulate gene expression by binding to partially complementary regions within the 3'UTR of their target genes. Computational methods play an important role in target prediction and assume that the miRNA "seed region" (nt 2 to 8) is required for functional targeting, but typically only identify ?80% of known bindings. Recent studies have highlighted a role for the entire miRNA, suggesting that a more flexible methodology is needed. We present a novel approach for miRNA target prediction based on Deep Learning (DL) which, rather than incorporating any knowledge (such as seed regions), investigates the entire miRNA and 3'TR mRNA nucleotides to learn a uninhibited set of feature descriptors related to the targeting process. We collected more than 150,000 experimentally validated homo sapiens miRNA:gene targets and cross referenced them with different CLIP-Seq, CLASH and iPAR-CLIP datasets to obtain ?20,000 validated miRNA:gene exact target sites. Using this data, we implemented and trained a deep neural network-composed of autoencoders and a feed-forward network-able to automatically learn features describing miRNA-mRNA interactions and assess functionality. Predictions were then refined using information such as site location or site accessibility energy. In a comparison using independent datasets, our DL approach consistently outperformed existing prediction methods, recognizing the seed region as a common feature in the targeting process, but also identifying the role of pairings outside this region. Thermodynamic analysis also suggests that site accessibility plays a role in targeting but that it cannot be used as a sole indicator for functionality. Data and source code available at: https://bitbucket.org/account/user/bipous/projects/MIRAW.
| S-EPMC6067737 | biostudies-literature
Project description:The blood-brain barrier plays a crucial role in regulating the passage of 98% of the compounds that enter the central nervous system (CNS). Compounds with high permeability must be identified to enable the synthesis of brain medications for the treatment of various brain diseases, such as Parkinson's, Alzheimer's, and brain tumors. Throughout the years, several models have been developed to solve this problem and have achieved acceptable accuracy scores in predicting compounds that penetrate the blood-brain barrier. However, predicting compounds with "low" permeability has been a challenging task. In this study, we present a deep learning (DL) classification model to predict blood-brain barrier permeability. The proposed model addresses the fundamental issues presented in former models: high dimensionality, class imbalances, and low specificity scores. We address these issues to enhance the high-dimensional, imbalanced dataset before developing the classification model: the imbalanced dataset is addressed using oversampling techniques and the high dimensionality using a non-linear dimensionality reduction technique known as kernel principal component analysis (KPCA). This technique transforms the high-dimensional dataset into a low-dimensional Euclidean space while retaining invaluable information. For the classification task, we developed an enhanced feed-forward deep learning model and a convolutional neural network model. In terms of specificity scores (i.e., predicting compounds with low permeability), the results obtained by the enhanced feed-forward deep learning model outperformed those obtained by other models in the literature that were developed using the same technique. In addition, the proposed convolutional neural network model surpassed models used in other studies in multiple accuracy measures, including overall accuracy and specificity. The proposed approach solves the problem inevitably faced with obtaining low specificity resulting in high false positive rate.
| S-EPMC8205267 | biostudies-literature
Project description:ObjectiveTo develop and evaluate deep learning (DL) risk assessment models for predicting pain progression in subjects with or at risk of knee osteoarthritis (OA).Materials and methodsThe incidence and progression cohorts of the Osteoarthritis Initiative, a multi-center longitudinal study involving 9348 knees in 4674 subjects with or at risk of knee OA that began in 2004 and is ongoing, were used to conduct this retrospective analysis. A subset of knees without and with pain progression (defined as a 9-point or greater increase in pain score between baseline and two or more follow-up time points over the first 48 months) was randomly stratified into training (4200 knees with a mean age of 61.0 years and 60% female) and hold-out testing (500 knees with a mean age of 60.8 years and 60% female) datasets. A DL model was developed to predict pain progression using baseline knee radiographs. An artificial neural network was used to develop a traditional risk assessment model to predict pain progression using demographic, clinical, and radiographic risk factors. A combined model was developed to combine demographic, clinical, and radiographic risk factors with DL analysis of baseline knee radiographs. Area under the curve (AUC) analysis was performed using the hold-out testing dataset to evaluate model performance.ResultsThe traditional model had an AUC of 0.692 (66.9% sensitivity and 64.1% specificity). The DL model had an AUC of 0.770 (76.7% sensitivity and 70.5% specificity), which was significantly higher (p < 0.001) than the traditional model. The combined model had an AUC of 0.807 (72.3% sensitivity and 80.9% specificity), which was significantly higher (p < 0.05) than the traditional and DL models.ConclusionsDL models using baseline knee radiographs had higher diagnostic performance for predicting pain progression than traditional models using demographic, clinical, and radiographic risk factors.
| S-EPMC9232386 | biostudies-literature
Project description:ObjectivesMany deep learning-based predictive models evaluate the waveforms of electrocardiograms (ECGs). Because deep learning-based models are data-driven, large and labeled biosignal datasets are required. Most individual researchers find it difficult to collect adequate training data. We suggest that transfer learning can be used to solve this problem and increase the effectiveness of biosignal analysis.MethodsWe applied the weights of a pretrained model to another model that performed a different task (i.e., transfer learning). We used 2,648,100 unlabeled 8.2-second-long samples of ECG II data to pretrain a convolutional autoencoder (CAE) and employed the CAE to classify 12 ECG rhythms within a dataset, which had 10,646 10-second-long 12-lead ECGs with 11 rhythm labels. We split the datasets into training and test datasets in an 8:2 ratio. To confirm that transfer learning was effective, we evaluated the performance of the classifier after the proposed transfer learning, random initialization, and two-dimensional transfer learning as the size of the training dataset was reduced. All experiments were repeated 10 times using a bootstrapping method. The CAE performance was evaluated by calculating the mean squared errors (MSEs) and that of the ECG rhythm classifier by deriving F1-scores.ResultsThe MSE of the CAE was 626.583. The mean F1-scores of the classifiers after bootstrapping of 100%, 50%, and 25% of the training dataset were 0.857, 0.843, and 0.835, respectively, when the proposed transfer learning was applied and 0.843, 0.831, and 0.543, respectively, after random initialization was applied.ConclusionsTransfer learning effectively overcomes the data shortages that can compromise ECG domain analysis by deep learning.
| S-EPMC7921576 | biostudies-literature