Protein-protein interaction network construction for cancer using a new L1/2-penalized Net-SVM model

Chai, Hua; Huang, Haihui; Jiang, Hong-Kun; Liang, Yong; Xia, Liang-Yong

doi:10.4238/gmr.15038794

Cited by 5 publications

(5 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The superiority of this method over the existing sequence-based methods will make it useful for the study of protein networks. Chai et al (69) built a new Net-SVM model which selected fewer but more relevant genes. This Net-SVM can be used to construct simple and informative PPI networks that are highly relevant to cancer.…”

Section: Cancer Gene/protein Interaction and Networkmentioning

confidence: 99%

Applications of Support Vector Machine (SVM) Learning in Cancer Genomics

2018

CGP

685

347

View full text Add to dashboard Cite

Abstract. Machine learning with maximization (support) of separating margin (vector), called support vector machine (SVM) learning, is a powerful classification tool that hasMachine learning (ML) "learns" a model from past data in order to predict future data (1). The key process is the learning which is one of the artificial intelligences. Many different statistical, probabilistic, and optimization techniques can be implemented as the learning methods such as the logistic regression, artificial neural networks (ANN), K-nearest neighbor (KNN), decision trees (DT) and Naive Bayes. There are two main types of ML learning -supervised learning and unsupervised learning. The supervised learning builds a model by learning from known classes (labeled training data). In contrast, unsupervised learning methods learn the common features from unknown class data (unlabeled training data).ML algorithms have been used for key feature training and recognition and for group classification. The strength of ML methods is it could detect hard-to-discern patterns from large, noisy or complex data sets. This capability is particularly well-suited to complex genomic data, especially in cancer studies. For example, ANN and DT have been used in cancer detection and diagnosis for nearly 20 years (2-3). The clinical implication of cancer heterogeneity and various cancer genomic data available motivate the applications of ML for cancer classification using genomic data.SVM learning is one of many ML methods. Compared to the other ML methods SVM is very powerful at recognizing subtle patterns in complex datasets (4). SVM can be used to recognize handwriting, recognize fraudulent credit cards, identify a speaker, as well as detect face (5). Cancer is a genetic disease where the genomic feature patterns or feature function patterns may represent the cancer subtypes, the outcome prognosis, drug benefit prediction, tumorigenesis drivers, or a tumor-specific biological process. Therefore, the Artificial Intelligence of SVM can help us in recognizing these patterns in a variety of applications. SVM ModelSVM is a powerful method for building a classifier. It aims to create a decision boundary between two classes that enables the prediction of labels from one or more feature vectors (6). This decision boundary, known as the hyperplane, is orientated in such a way that it is as far as 41

show abstract

Section: Cancer Gene/protein Interaction and Networkmentioning

confidence: 99%

Applications of Support Vector Machine (SVM) Learning in Cancer Genomics

2018

CGP

685

347

View full text Add to dashboard Cite

show abstract

“…The smaller number of seedlings produced in plastic bags, together with the greater variation amplitude of height (25.60 cm), may have negatively affected the predictive performance of PDF Logistic 2P, Log-logistic 2P, Gamma 2P, Normal and Log-normal. It is important to emphasize that other statistical approaches, such as the use of artificial intelligence and regularized regressions (Binoti et al, 2013;Castro et al, 2013;Binoti et al, 2014;Kadyrova and Pavlova, 2014;Diamantopoulou et al, 2015;Chai et al, 2016), can be applied in order to improve predictive quality in complex database modeling. The most common regularization methods for machine learning focused on regression problems are the Ridge and Least Absolute Shrinkage and Selection Operator (LASSO) (Chai et al, 2016).…”

Section: Discussionmentioning

confidence: 99%

“…It is important to emphasize that other statistical approaches, such as the use of artificial intelligence and regularized regressions (Binoti et al, 2013;Castro et al, 2013;Binoti et al, 2014;Kadyrova and Pavlova, 2014;Diamantopoulou et al, 2015;Chai et al, 2016), can be applied in order to improve predictive quality in complex database modeling. The most common regularization methods for machine learning focused on regression problems are the Ridge and Least Absolute Shrinkage and Selection Operator (LASSO) (Chai et al, 2016). Biological network-regularized logistic models are examples that have been extensively used in the genomic area (Zhang et al, 2013;Huang et al, 2015;Chai et al, 2016;Huang et al, 2016;Kang et al, 2017), but its application has not yet been found in Brazilian silviculture.…”

Section: Discussionmentioning

confidence: 99%

“…The most common regularization methods for machine learning focused on regression problems are the Ridge and Least Absolute Shrinkage and Selection Operator (LASSO) (Chai et al, 2016). Biological network-regularized logistic models are examples that have been extensively used in the genomic area (Zhang et al, 2013;Huang et al, 2015;Chai et al, 2016;Huang et al, 2016;Kang et al, 2017), but its application has not yet been found in Brazilian silviculture.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Modeling of hypsometric distribution of Handroanthus heptaphyllus seedlings in different containers

Lafetá¹,

Rodrigues²,

Penido³

et al. 2018

Afr. J. Agric. Res.

View full text Add to dashboard Cite

Hypsometric information of seedlings allows greater assertiveness of silvicultural decisions in nurseries. This study aimed to evaluate the efficiency of different Probability Density Functions (PDF) to estimate and compare the height distribution of Handroanthus heptaphyllus seedlings in different cultivation containers. One hundred and thirty one seedlings were produce in two types of containers; 81 units in tubes and the remaining in plastic bags. The census was realized at 122-day old, measuring total height of the seedlings using millimeter ruler. The seedlings did not present a tender or brittle caulinar system. Data were grouped into biometric classes with regular intervals of 2.5 cm in height. Seven PDF were adjusted using maximum likelihood method and the one with the best predictive performance was selected to identify the statistical equality between the distributions estimated for the seedling size in each container. The order of predictive efficiency of PDF was distinct between recipients. The Weibull of two parameters function can be used to model the height distribution of H. heptaphyllus seedlings at 122-day old, produced in plastic bags and tubes. The hypsometric distribution was different between containers.

show abstract

“…Although not always having the best accuracy, the overall performance of SVM models regardless of the configuration, are modest at the very least. SVM performed well, even with low sample size and high-dimensional dataset like in Chai et al [25] and Matsumoto et al [24]. Similarly, in Durgesh and Lekha [23], SVM can have different kernel functions for various characteristics of a dataset.…”

Section: Literature Analysismentioning

confidence: 98%

A comparative performance of breast cancer classification using hyper-parameterized machine learning models

Mayce¹,

Hernandez²

2021

IJATEE

View full text Add to dashboard Cite

Cancer is the result of the outgrowth of a clonal population of cells from bodily tissue [1]. It has been acknowledged that early oncogenesis, also known as carcinogenesis, can be determined by cell-intrinsic features. A way to determine is by illustrating these features of both cancer cells and tumors , which are the ability to provide their growth signals, disregard to growth-inhibitory signals, eluding of apoptosis (programmed cell death), unlimited replication, sustained vascularization and malignancy, invasion of tissues through basement membranes and capillary walls. Cancer is considered to have three developmental phases: initiation, promotion, and progression. In the initiation phase, genomic changes such as point mutations, amplification and gene deletion, and rearrangements in the chromosomes within the cancer cell occur. *Author for correspondence These initiated cells expand clonally and survive promoting tumor development in the promotion phase. In the progression phase, significant tumor growth and metastasis are encompassed. Accumulation of genetic lesions also plays a role in the development of cancer. Inarguably, this phenomenon is present in the Initiation phase, but it may also be present in the promotion and progression phase. In the accumulation of genetic lesions, inactivation of tumor suppressor genes also occurs which provides the cells certain properties that are needed for tumorigenesis. Many types of research have established that both cancer cells and noncancer cells with a significant amount are the recipes for the heterocellular tumor are involved in tumor development and malignancy [2, 3].In most cases, cancer derives its name after the body part in which it emanated therefore, breast cancer is the term for the unstable growth of cells that emanated in the breast tissue. The growth results in a Re search Article

show abstract

Protein-protein interaction network construction for cancer using a new L1/2-penalized Net-SVM model

Cited by 5 publications

References 29 publications

Applications of Support Vector Machine (SVM) Learning in Cancer Genomics

Applications of Support Vector Machine (SVM) Learning in Cancer Genomics

Modeling of hypsometric distribution of Handroanthus heptaphyllus seedlings in different containers

A comparative performance of breast cancer classification using hyper-parameterized machine learning models

Contact Info

Product

Resources

About