A Novel Algorithm for the Precise Calculation of the Maximal Information Coefficient

Zhang, Yi; Jia, Shili; Huang, Haiyun; Qiu, Jiqing; Zhou, Chang-Jie

doi:10.1038/srep06662

Cited by 54 publications

(32 citation statements)

References 22 publications

(37 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The MIC method is able to measure the association between two signals regardless of whether the data are linear or show any type of nonlinearity, but it requires a fairly long time course to obtain a precise estimation of the mutual information (Reshef et al ., ; Zhang et al ., ). That is why we adopted a large sliding window (T = 500) for the analysis of stationarity.…”

Section: Discussionmentioning

confidence: 97%

Different interaction modes for the default mode network revealed by resting state functional magnetic resonance imaging

Zuo

Song

Fan

et al. 2015

Eur J of Neuroscience

View full text Add to dashboard Cite

The default mode network (DMN), which, in the resting state, is in charge of both the brain's intrinsic mentation and its reflexive responses to external stimuli, is recognized as an essential network in the human brain. These two roles of mentation and reflexive response recruit the DMN nodes and other task networks differently. Existing research has revealed that the interactions inside the DMN (between nodes within the DMN) and outside the DMN (between nodes in the DMN and nodes in task networks) have different modes, in terms of both strength and timing. These findings raise interesting questions. For example, are the internal and external interactions of the DMN equally linear or nonlinear? This study examined these interaction patterns using datasets from the Human Connectome Project. A maximal information-based nonparametric exploration statistics strategy was utilized to characterize the full correlations, and the Pearson correlation was used to capture the linear component of the full correlations. We then contrasted the level of linearity/nonlinearity with respect to the internal and external interactions of the DMN. After a brain-wide exploration, we found that the interactions between the DMN and the sensorimotor-related networks (including the sensorimotor, sensory association, and integration areas) showed more nonlinearity, whereas those between the intra-DMN nodes were similarly less nonlinear. These findings may provide a clue for understanding the underlying neuronal principles of the internal and external roles of the DMN.

show abstract

Section: Discussionmentioning

confidence: 97%

Different interaction modes for the default mode network revealed by resting state functional magnetic resonance imaging

Zuo

Song

Fan

et al. 2015

Eur J of Neuroscience

View full text Add to dashboard Cite

show abstract

“…The maximal information coefficient (MIC) was proposed to capture a wide range of associations of two variables, in both linear and nonlinear relationships (Reshef et al, 2011). Owing to its generality, MIC is becoming widely accepted in scientific research (Zhang et al, 2014), and is also used to analyze large biological datasets (Rau et al, 2013;Wang and Zhao, 2015;Wang et al, 2016). However, even with MIC, it is difficult to identify genes with a nonlinear expression pattern as the genes giving the most strongly supported hits are linearly expressed.…”

Section: Introductionmentioning

confidence: 99%

A Novel Method to Efficiently Highlight Nonlinearly Expressed Genes

et al. 2020

View full text Add to dashboard Cite

For precision medicine, there is a need to identify genes that accurately distinguish the physiological state or response to a particular therapy, but this can be challenging. Many methods of analyzing differential expression have been established and applied to this problem, such as t-test, edgeR, and DEseq2. A common feature of these methods is their focus on a linear relationship (differential expression) between gene expression and phenotype. However, they may overlook nonlinear relationships due to various factors, such as the degree of disease progression, sex, age, ethnicity, and environmental factors. Maximal information coefficient (MIC) was proposed to capture a wide range of associations of two variables in both linear and nonlinear relationships. However, with MIC it is difficult to highlight genes with nonlinear expression patterns as the genes giving the most strongly supported hits are linearly expressed, especially for noisy data. It is thus important to also efficiently identify nonlinearly expressed genes in order to unravel the molecular basis of disease and to reveal new therapeutic targets. We propose a novel nonlinearity measure called normalized differential correlation (NDC) to efficiently highlight nonlinearly expressed genes in transcriptome datasets. Validation using six real-world cancer datasets revealed that the NDC method could highlight nonlinearly expressed genes that could not be highlighted by t-test, MIC, edgeR, and DEseq2, although MIC could capture nonlinear correlations. The classification accuracy indicated that analysis of these genes could adequately distinguish cancer and paracarcinoma tissue samples. Furthermore, the results of biological interpretation of the identified genes suggested that some of them were involved in key functional pathways associated with cancer progression and metastasis. All of this evidence suggests that these nonlinearly expressed genes may play a central role in regulating cancer progression.

show abstract

“…Moreover, a high value of MIC indicates a high correlation between the corresponding variables, whereas MIC ¼ 0 indicates that two corresponding variables are independent (Ge et al, 2016). The MIC of two variables x i and x j is defined as follows (Zhang, Jia, Huang, Qiu, & Zhou, 2014):…”

Section: Correlated and Weakly Correlated Variable Divisionmentioning

confidence: 99%

“…And pðx k i Þ denotes the probability of variable x i is x k i (Reshef et al, 2011;Zhang et al, 2014).…”

Section: Correlated and Weakly Correlated Variable Divisionmentioning

confidence: 99%

Correlated and weakly correlated fault detection based on variable division and ICA

Zhou

Pan

et al. 2017

Computers & Industrial Engineering

View full text Add to dashboard Cite

A Novel Algorithm for the Precise Calculation of the Maximal Information Coefficient

Cited by 54 publications

References 22 publications

Different interaction modes for the default mode network revealed by resting state functional magnetic resonance imaging

Different interaction modes for the default mode network revealed by resting state functional magnetic resonance imaging

A Novel Method to Efficiently Highlight Nonlinearly Expressed Genes

Correlated and weakly correlated fault detection based on variable division and ICA

Contact Info

Product

Resources

About