scholarly journals HPOLabeler: improving prediction of human protein–phenotype associations by learning to rank

2020 ◽  
Vol 36 (14) ◽  
pp. 4180-4188
Author(s):  
Lizhi Liu ◽  
Xiaodi Huang ◽  
Hiroshi Mamitsuka ◽  
Shanfeng Zhu
Keyword(s):  
Large Scale ◽  
Nearest Neighbor ◽  
Learning To Rank ◽  
Supplementary Information ◽  
Human Phenotype ◽  
Protein Protein Interaction ◽  
Ppi Networks ◽  
Naive Method ◽  
Temporal Validation ◽  
Ranking Problems

Abstract Motivation Annotating human proteins by abnormal phenotypes has become an important topic. Human Phenotype Ontology (HPO) is a standardized vocabulary of phenotypic abnormalities encountered in human diseases. As of November 2019, only <4000 proteins have been annotated with HPO. Thus, a computational approach for accurately predicting protein–HPO associations would be important, whereas no methods have outperformed a simple Naive approach in the second Critical Assessment of Functional Annotation, 2013–2014 (CAFA2). Results We present HPOLabeler, which is able to use a wide variety of evidence, such as protein–protein interaction (PPI) networks, Gene Ontology, InterPro, trigram frequency and HPO term frequency, in the framework of learning to rank (LTR). LTR has been proved to be powerful for solving large-scale, multi-label ranking problems in bioinformatics. Given an input protein, LTR outputs the ranked list of HPO terms from a series of input scores given to the candidate HPO terms by component learning models (logistic regression, nearest neighbor and a Naive method), which are trained from given multiple evidence. We empirically evaluate HPOLabeler extensively through mainly two experiments of cross validation and temporal validation, for which HPOLabeler significantly outperformed all component models and competing methods including the current state-of-the-art method. We further found that (i) PPI is most informative for prediction among diverse data sources and (ii) low prediction performance of temporal validation might be caused by incomplete annotation of new proteins. Availability and implementation http://issubmission.sjtu.edu.cn/hpolabeler/. Contact [email protected] Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Short loop functional commonality identified in leukaemia proteome highlights crucial protein sub-networks

2021 ◽  
Vol 3 (1) ◽  
Author(s):  
Sun Sook Chung ◽  
Joseph C F Ng ◽  
Anna Laddach ◽  
N Shaun B Thomas ◽  
Franca Fraternali
Keyword(s):  
Protein Interactions ◽  
Large Scale ◽  
Interaction Network ◽  
Protein Protein Interactions ◽  
Protein Protein Interaction ◽  
Ppi Networks ◽  
Short Loop ◽  
New Strategy ◽  
Loop Network ◽  
Protein Protein Interaction Network

Abstract Direct drug targeting of mutated proteins in cancer is not always possible and efficacy can be nullified by compensating protein–protein interactions (PPIs). Here, we establish an in silico pipeline to identify specific PPI sub-networks containing mutated proteins as potential targets, which we apply to mutation data of four different leukaemias. Our method is based on extracting cyclic interactions of a small number of proteins topologically and functionally linked in the Protein–Protein Interaction Network (PPIN), which we call short loop network motifs (SLM). We uncover a new property of PPINs named ‘short loop commonality’ to measure indirect PPIs occurring via common SLM interactions. This detects ‘modules’ of PPI networks enriched with annotated biological functions of proteins containing mutation hotspots, exemplified by FLT3 and other receptor tyrosine kinase proteins. We further identify functional dependency or mutual exclusivity of short loop commonality pairs in large-scale cellular CRISPR–Cas9 knockout screening data. Our pipeline provides a new strategy for identifying new therapeutic targets for drug discovery.

scholarly journals BERTMeSH: deep contextual representation learning for large-scale high-performance MeSH indexing with full text

2020 ◽  
Author(s):  
Ronghui You ◽  
Yuxuan Liu ◽  
Hiroshi Mamitsuka ◽  
Shanfeng Zhu
Keyword(s):  
Full Text ◽  
High Performance ◽  
Large Scale ◽  
Learning Strategy ◽  
Learning To Rank ◽  
Representation Learning ◽  
Supplementary Information ◽  
Medical Subject Headings ◽  
The Difference ◽  
Contextual Representation

Abstract Motivation With the rapid increase of biomedical articles, large-scale automatic Medical Subject Headings (MeSH) indexing has become increasingly important. FullMeSH, the only method for large-scale MeSH indexing with full text, suffers from three major drawbacks: FullMeSH (i) uses Learning To Rank, which is time-consuming, (ii) can capture some pre-defined sections only in full text and (iii) ignores the whole MEDLINE database. Results We propose a computationally lighter, full text and deep-learning-based MeSH indexing method, BERTMeSH, which is flexible for section organization in full text. BERTMeSH has two technologies: (i) the state-of-the-art pre-trained deep contextual representation, Bidirectional Encoder Representations from Transformers (BERT), which makes BERTMeSH capture deep semantics of full text. (ii) A transfer learning strategy for using both full text in PubMed Central (PMC) and title and abstract (only and no full text) in MEDLINE, to take advantages of both. In our experiments, BERTMeSH was pre-trained with 3 million MEDLINE citations and trained on ∼1.5 million full texts in PMC. BERTMeSH outperformed various cutting-edge baselines. For example, for 20 K test articles of PMC, BERTMeSH achieved a Micro F-measure of 69.2%, which was 6.3% higher than FullMeSH with the difference being statistically significant. Also prediction of 20 K test articles needed 5 min by BERTMeSH, while it took more than 10 h by FullMeSH, proving the computational efficiency of BERTMeSH. Supplementary information Supplementary data are available at Bioinformatics online

scholarly journals Protein–protein interaction sites prediction by ensemble random forests with synthetic minority oversampling technique

2018 ◽  
Vol 35 (14) ◽  
pp. 2395-2402 ◽  
Author(s):  
Xiaoying Wang ◽  
Bin Yu ◽  
Anjun Ma ◽  
Cheng Chen ◽  
Bingqiang Liu ◽  
...  
Keyword(s):  
Ensemble Learning ◽  
Protein Interaction ◽  
Feature Selection Method ◽  
Feature Space ◽  
Supplementary Information ◽  
Sequence Profile ◽  
Protein Protein Interaction ◽  
Interaction Sites ◽  
Imbalance Problem ◽  
Ppi Networks

Abstract Motivation The prediction of protein–protein interaction (PPI) sites is a key to mutation design, catalytic reaction and the reconstruction of PPI networks. It is a challenging task considering the significant abundant sequences and the imbalance issue in samples. Results A new ensemble learning-based method, Ensemble Learning of synthetic minority oversampling technique (SMOTE) for Unbalancing samples and RF algorithm (EL-SMURF), was proposed for PPI sites prediction in this study. The sequence profile feature and the residue evolution rates were combined for feature extraction of neighboring residues using a sliding window, and the SMOTE was applied to oversample interface residues in the feature space for the imbalance problem. The Multi-dimensional Scaling feature selection method was implemented to reduce feature redundancy and subset selection. Finally, the Random Forest classifiers were applied to build the ensemble learning model, and the optimal feature vectors were inserted into EL-SMURF to predict PPI sites. The performance validation of EL-SMURF on two independent validation datasets showed 77.1% and 77.7% accuracy, which were 6.2–15.7% and 6.1–18.9% higher than the other existing tools, respectively. Availability and implementation The source codes and data used in this study are publicly available at http://github.com/QUST-AIBBDRC/EL-SMURF/. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Systems Biology Inferring edge function in protein-protein interaction networks

2018 ◽  
Author(s):  
Daniel Esposito ◽  
Joseph Cursons ◽  
Melissa Davis
Keyword(s):  
Protein Interaction ◽  
Supplementary Information ◽  
Human Interactome ◽  
High Coverage ◽  
Protein Protein Interaction ◽  
Cellular Processes ◽  
Regulatory Processes ◽  
Ppi Networks ◽  
Biological Phenomena ◽  
Biological Insight

AbstractMotivation: Post-translational modifications (PTMs) regulate many key cellular processes. Numerous studies have linked the topology of protein-protein interaction (PPI) networks to many biological phenomena such as key regulatory processes and disease. However, these methods fail to give insight in the functional nature of these interactions. On the other hand, pathways are commonly used to gain biological insight into the function of PPIs in the context of cascading interactions, sacrificing the coverage of networks for rich functional annotations on each PPI. We present a machine learning approach that uses Gene Ontology, InterPro and Pfam annotations to infer the edge functions in PPI networks, allowing us to combine the high coverage of networks with the information richness of pathways.Results: An ensemble method with a combination Logistic Regression and Random Forest classifiers trained on a high-quality set of annotated interactions, with a total of 18 unique labels, achieves high a average F1 score 0.88 despite not taking advantage of multi-label dependencies. When applied to the human interactome, our method confidently classifies 62% of interactions at a probability of 0.7 or higher.Availability: Software and data are available at https://github.com/DavisLaboratory/pyPPIContact:[email protected] information: Supplementary data are available at Bioinformatics online.

scholarly journals FullMeSH: improving large-scale MeSH indexing with full text

2019 ◽  
Vol 36 (5) ◽  
pp. 1533-1541
Author(s):  
Suyang Dai ◽  
Ronghui You ◽  
Zhiyong Lu ◽  
Xiaodi Huang ◽  
Hiroshi Mamitsuka ◽  
...  
Keyword(s):  
Full Text ◽  
Large Scale ◽  
Learning To Rank ◽  
Biomedical Literature ◽  
Supplementary Information ◽  
Limited Information ◽  
Pubmed Central ◽  
Average Improvement ◽  
Indexing Method ◽  
Overall Performance

Abstract Motivation With the rapidly growing biomedical literature, automatically indexing biomedical articles by Medical Subject Heading (MeSH), namely MeSH indexing, has become increasingly important for facilitating hypothesis generation and knowledge discovery. Over the past years, many large-scale MeSH indexing approaches have been proposed, such as Medical Text Indexer, MeSHLabeler, DeepMeSH and MeSHProbeNet. However, the performance of these methods is hampered by using limited information, i.e. only the title and abstract of biomedical articles. Results We propose FullMeSH, a large-scale MeSH indexing method taking advantage of the recent increase in the availability of full text articles. Compared to DeepMeSH and other state-of-the-art methods, FullMeSH has three novelties: (i) Instead of using a full text as a whole, FullMeSH segments it into several sections with their normalized titles in order to distinguish their contributions to the overall performance. (ii) FullMeSH integrates the evidence from different sections in a ‘learning to rank’ framework by combining the sparse and deep semantic representations. (iii) FullMeSH trains an Attention-based Convolutional Neural Network for each section, which achieves better performance on infrequent MeSH headings. FullMeSH has been developed and empirically trained on the entire set of 1.4 million full-text articles in the PubMed Central Open Access subset. It achieved a Micro F-measure of 66.76% on a test set of 10 000 articles, which was 3.3% and 6.4% higher than DeepMeSH and MeSHLabeler, respectively. Furthermore, FullMeSH demonstrated an average improvement of 4.7% over DeepMeSH for indexing Check Tags, a set of most frequently indexed MeSH headings. Availability and implementation The software is available upon request. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Hybrid approach for disease comorbidity and disease gene prediction using heterogeneous dataset

2021 ◽  
Vol 11 (6) ◽  
pp. 5240
Author(s):  
Lakshmi K. S. ◽  
Vadivu G.
Keyword(s):  
Large Scale ◽  
Disease Gene ◽  
Gene Prediction ◽  
Hybrid Approach ◽  
Biological Data ◽  
High Throughput Analysis ◽  
Human Phenotype ◽  
Protein Protein Interaction ◽  
Using Data ◽  
Scale Integration

High throughput analysis and large scale integration of biological data led to leading researches in the field of bioinformatics. Recent years witnessed the development of various methods for disease associated gene prediction and disease comorbidity predictions. Most of the existing techniques use network-based approaches and similarity-based approaches for these predictions. Even though network-based approaches have better performance, these methods rely on text data from OMIM records and PubMed abstracts. In this method, a novel algorithm (HDCDGP) is proposed for disease comorbidity prediction and disease associated gene prediction. Disease comorbidity network and disease gene network were constructed using data from gene ontology (GO), human phenotype ontology (HPO), protein-protein interaction (PPI) and pathway dataset. Modified random walk restart algorithm was applied on these networks for extracting novel disease-gene associations. Experimental results showed that the hybrid approach has better performance compared to existing systems with an overall accuracy around 85%.

A novel method for data fusion over entity-relation graphs and its application to protein–protein interaction prediction

2021 ◽  
Author(s):  
Daniele Raimondi ◽  
Jaak Simm ◽  
Adam Arany ◽  
Yves Moreau
Keyword(s):  
Data Fusion ◽  
Protein Interactions ◽  
Protein Function ◽  
Scientific Progress ◽  
Supplementary Information ◽  
Sources Of Information ◽  
Protein Protein Interaction ◽  
Ppi Networks ◽  
Proteome Level ◽  
Clear Shift

Abstract Motivation Modern bioinformatics is facing increasingly complex problems to solve, and we are indeed rapidly approaching an era in which the ability to seamlessly integrate heterogeneous sources of information will be crucial for the scientific progress. Here, we present a novel non-linear data fusion framework that generalizes the conventional matrix factorization paradigm allowing inference over arbitrary entity-relation graphs, and we applied it to the prediction of protein–protein interactions (PPIs). Improving our knowledge of PPI networks at the proteome scale is indeed crucial to understand protein function, physiological and disease states and cell life in general. Results We devised three data fusion-based models for the proteome-level prediction of PPIs, and we show that our method outperforms state of the art approaches on common benchmarks. Moreover, we investigate its predictions on newly published PPIs, showing that this new data has a clear shift in its underlying distributions and we thus train and test our models on this extended dataset. Supplementary information Supplementary data are available at Bioinformatics online.

A High Performance Parallel Ranking SVM with OpenCL on Multi-core and Many-core Platforms

2019 ◽  
Vol 11 (1) ◽  
pp. 17-28 ◽  
Author(s):  
Huming Zhu ◽  
Pei Li ◽  
Peng Zhang ◽  
Zheng Luo
Keyword(s):  
Parallel Programming ◽  
High Performance ◽  
Large Scale ◽  
Graphics Processing Unit ◽  
Learning To Rank ◽  
Support Vector ◽  
Processing Unit ◽  
Ranking Svm ◽  
Ranking Problems ◽  
Many Core

A ranking support vector machine (RSVM) is a typical pairwise method of learning to rank, which is effective in ranking problems. However, the training speed of RSVMs are not satisfactory, especially when solving large-scale data ranking problems. Recent years, many-core processing units (graphics processing unit (GPU), Many Integrated Core (MIC)) and multi-core processing units have exhibited huge superiority in the parallel computing domain. With the support of hardware, parallel programming develops rapidly. Open Computing Language (OpenCL) and Open Multi-Processing (OpenMP) are two of popular parallel programming interfaces. The authors present two high-performance parallel implementations of RSVM, an OpenCL version implemented on multi-core and many-core platforms, and an OpenMP version implemented on multi-core platform. The experimental results show that the OpenCL version parallel RSVM achieved considerable speedup on Intel MIC 7110P, NVIDIA Tesla K20M and Intel Xeon E5-2692v2, and it also shows good portability.

scholarly journals HPOAnnotator: improving large-scale prediction of HPO annotations by low-rank approximation with HPO semantic similarities and multiple PPI networks

2019 ◽  
Vol 12 (S10) ◽  
Author(s):  
Junning Gao ◽  
Lizhi Liu ◽  
Shuwei Yao ◽  
Xiaodi Huang ◽  
Hiroshi Mamitsuka ◽  
...  
Keyword(s):  
Hierarchical Structure ◽  
Matrix Factorization ◽  
Large Scale ◽  
Low Rank ◽  
Superior Performance ◽  
Low Rank Approximation ◽  
Human Phenotype ◽  
Ppi Networks ◽  
The Hierarchical Structure ◽  
Rank Approximation

Abstract Background As a standardized vocabulary of phenotypic abnormalities associated with human diseases, the Human Phenotype Ontology (HPO) has been widely used by researchers to annotate phenotypes of genes/proteins. For saving the cost and time spent on experiments, many computational approaches have been proposed. They are able to alleviate the problem to some extent, but their performances are still far from satisfactory. Method For inferring large-scale protein-phenotype associations, we propose HPOAnnotator that incorporates multiple Protein-Protein Interaction (PPI) information and the hierarchical structure of HPO. Specifically, we use a dual graph to regularize Non-negative Matrix Factorization (NMF) in a way that the information from different sources can be seamlessly integrated. In essence, HPOAnnotator solves the sparsity problem of a protein-phenotype association matrix by using a low-rank approximation. Results By combining the hierarchical structure of HPO and co-annotations of proteins, our model can well capture the HPO semantic similarities. Moreover, graph Laplacian regularizations are imposed in the latent space so as to utilize multiple PPI networks. The performance of HPOAnnotator has been validated under cross-validation and independent test. Experimental results have shown that HPOAnnotator outperforms the competing methods significantly. Conclusions Through extensive comparisons with the state-of-the-art methods, we conclude that the proposed HPOAnnotator is able to achieve the superior performance as a result of using a low-rank approximation with a graph regularization. It is promising in that our approach can be considered as a starting point to study more efficient matrix factorization-based algorithms.

scholarly journals PICKLE 3.0: enriching the human meta-database with the mouse protein interactome extended via mouse–human orthology

2020 ◽  
Author(s):  
Georgios N Dimitrakopoulos ◽  
Maria I Klapa ◽  
Nicholas K Moschonas
Keyword(s):  
Mouse Genome Informatics ◽  
Supplementary Information ◽  
Protein Protein Interaction ◽  
Ppi Networks ◽  
Mouse Protein ◽  
Protein Interactome ◽  
Primary Mouse ◽  
Relational Scheme ◽  
Human Ortholog ◽  
Experimental Mouse

Abstract Summary The PICKLE 3.0 upgrade refers to the enrichment of this human protein–protein interaction (PPI) meta-database with the mouse protein interactome. Experimental PPI data between mouse genetic entities are rather limited; however, they are substantially complemented by PPIs between mouse and human genetic entities. The relational scheme of PICKLE 3.0 has been amended to exploit the Mouse Genome Informatics mouse–human ortholog gene pair collection, enabling (i) the extension through orthology of the mouse interactome with potentially valid PPIs between mouse entities based on the experimental PPIs between mouse and human entities and (ii) the comparison between mouse and human PPI networks. Interestingly, 43.5% of the experimental mouse PPIs lacks a corresponding by orthology PPI in human, an inconsistency in need of further investigation. Overall, as primary mouse PPI datasets show a considerably limited overlap, PICKLE 3.0 provides a unique comprehensive representation of the mouse protein interactome. Availability and implementation PICKLE can be queried and downloaded at http://www.pickle.gr. Supplementary information Supplementary data are available at Bioinformatics online.

Sign in / Sign up

Export Citation Format

Share Document