Integrating Multiple Data Sources for Combinatorial Marker Discovery: A Study in Tumorigenesis

Bandyopadhyay, Sanghamitra; Mallik, Saurav

doi:10.1109/tcbb.2016.2636207

Cited by 23 publications

(3 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mukhopadhay et al [18] proposed a particleswarm optimization based gene-marker discovery technique. Building on the productive research for single marker discovery, now investigators aim to find combinatorial gene-based markers for disease [45][46][47][48][49][50].…”

Section: Literature Reviewmentioning

confidence: 99%

Towards integrated oncogenic marker recognition through mutual information‐based statistically significant feature extraction: an association rule mining based study on cancer expression and methylation profiles

Mallik

Zhao

2017

Quant. Biol.

Self Cite

View full text Add to dashboard Cite

Background Marker detection is an important task in complex disease studies. Here we provide an association rule mining (ARM) based approach for identifying integrated markers through mutual information (MI) based statistically significant feature extraction, and apply it to acute myeloid leukemia (AML) and prostate carcinoma (PC) gene expression and methylation profiles. Methods We first collect the genes having both expression and methylation values in AML as well as PC. Next, we run Jarque-Bera normality test on the expression/methylation data to divide the whole dataset into two parts: one that ollows normal distribution and the other that does not follow normal distribution. Thus, we have now four parts of the dataset: normally distributed expression data, normally distributed methylation data, non-normally distributed expression data, and non-normally distributed methylated data. A feature-extraction technique, “mRMR” is then utilized on each part. This results in a list of top-ranked genes. Next, we apply Welch t-test (parametric test) and Shrink t-test (non-parametric test) on the expression/methylation data for the top selected normally distributed genes and non-normally distributed genes, respectively. We then use a recent weighted ARM method, “RANWAR” to combine all/specific resultant genes to generate top oncogenic rules along with respective integrated markers. Finally, we perform literature search as well as KEGG pathway and Gene-Ontology (GO) analyses using Enrichr database for in silico validation of the prioritized oncogenes as the markers and labeling the markers as existing or novel. Results The novel markers of AML are {ABCB11↑∪KRT17↓} (i.e., ABCB11 as up-regulated, & KRT17 as down-regulated), and {AP1S1-∪KRT17↓∪NEIL2-∪DYDC1↓}) (i.e., AP1S1 and NEIL2 both as hypo-methylated, & KRT17 and DYDC1 both as down-regulated). The novel marker of PC is {UBIAD1¶∪APBA2‡∪C4orf31‡} (i.e., UBIAD1 as up-regulated and hypo-methylated, & APBA2 and C4orf31 both as down-regulated and hyper-methylated). Conclusion The identified novel markers might have critical roles in AML as well as PC. The approach can be applied to other complex disease.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Towards integrated oncogenic marker recognition through mutual information‐based statistically significant feature extraction: an association rule mining based study on cancer expression and methylation profiles

Mallik

Zhao

2017

Quant. Biol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this upcoming era, the multi-platform integration approach has been applied to accomplish various important tasks, such as signature/bio-marker detection, disease classification, and gene clustering. Prior research works in bio-marker discovery ( Bandyopadhyay and Mallik, 2016 ; Kandimalla et al, 2022 ), classification ( Henry et al, 2014 ; Maulik et al, 2015 ; Zhang and Kuster, 2019) , and clustering ( Wang and Gu, 2016 ) have improved the promising performance of multi-modal integration approaches. Nevertheless, the outcomes of such approaches are not always satisfactory.…”

Section: Introductionmentioning

confidence: 99%

3PNMF-MKL: A non-negative matrix factorization-based multiple kernel learning method for multi-modal data integration and its application to gene signature detection

et al. 2023

Self Cite

View full text Add to dashboard Cite

In this current era, biomedical big data handling is a challenging task. Interestingly, the integration of multi-modal data, followed by significant feature mining (gene signature detection), becomes a daunting task. Remembering this, here, we proposed a novel framework, namely, three-factor penalized, non-negative matrix factorization-based multiple kernel learning with soft margin hinge loss (3PNMF-MKL) for multi-modal data integration, followed by gene signature detection. In brief, limma, employing the empirical Bayes statistics, was initially applied to each individual molecular profile, and the statistically significant features were extracted, which was followed by the three-factor penalized non-negative matrix factorization method used for data/matrix fusion using the reduced feature sets. Multiple kernel learning models with soft margin hinge loss had been deployed to estimate average accuracy scores and the area under the curve (AUC). Gene modules had been identified by the consecutive analysis of average linkage clustering and dynamic tree cut. The best module containing the highest correlation was considered the potential gene signature. We utilized an acute myeloid leukemia cancer dataset from The Cancer Genome Atlas (TCGA) repository containing five molecular profiles. Our algorithm generated a 50-gene signature that achieved a high classification AUC score (viz., 0.827). We explored the functions of signature genes using pathway and Gene Ontology (GO) databases. Our method outperformed the state-of-the-art methods in terms of computing AUC. Furthermore, we included some comparative studies with other related methods to enhance the acceptability of our method. Finally, it can be notified that our algorithm can be applied to any multi-modal dataset for data integration, followed by gene module discovery.

show abstract

“…A microarray (Bandyopadhyay et al, 2014;Bandyopadhyay and Mallik, 2016) has been widely used to measure a large number of genes for determining differences between two groups (e.g., cases versus control samples), including gene expression profile, methylation, and genotype-based association studies. Methylation of cytosine (Navarro et al, 2012) changes the structure of DNA by introducing a methyl group (−CH3) at the carbon5 position of cytosine without altering the underlying DNA sequences.…”

Section: Introductionmentioning

confidence: 99%

Optimal ranking and directional signature classification using the integral strategy of multi-objective optimization-based association rule mining of multi-omics data

Mallik

Seth

et al. 2023

Front. Bioinform.

Self Cite

View full text Add to dashboard Cite

Introduction: Association rule mining (ARM) is a powerful tool for exploring the informative relationships among multiple items (genes) in any dataset. The main problem of ARM is that it generates many rules containing different rule-informative values, which becomes a challenge for the user to choose the effective rules. In addition, few works have been performed on the integration of multiple biological datasets and variable cutoff values in ARM.Methods: To solve all these problems, in this article, we developed a novel framework MOOVARM (multi-objective optimized variable cutoff-based association rule mining) for multi-omics profiles.Results: In this regard, we identified the positive ideal solution (PIS), which maximized the profit and minimized the loss, and negative ideal solution (NIS), which minimized the profit and maximized the loss for all gene sets (item sets), belonging to each extracted rule. Thereafter, we computed the distance (d +) from PIS and distance (d −) from NIS for each gene set or product. These two distances played an important role in determining the optimized associations among various pairs of genes in the multi-omics dataset. We then globally estimated the relative closeness to PIS for ranking the gene sets. When the relative closeness score of the rule is greater than or equal to the pre-defined threshold value, the rule can be considered a final resultant rule. Moreover, MOOVARM evaluated the relative score of the rule based on the status of all genes instead of individual genes.Conclusions:MOOVARM produced the final rank of the extracted (multi-objective optimized) rules of correlated genes which had better disease classification than the state-of-the-art algorithms on gene signature identification.

show abstract

Integrating Multiple Data Sources for Combinatorial Marker Discovery: A Study in Tumorigenesis

Cited by 23 publications

References 63 publications

Towards integrated oncogenic marker recognition through mutual information‐based statistically significant feature extraction: an association rule mining based study on cancer expression and methylation profiles

Towards integrated oncogenic marker recognition through mutual information‐based statistically significant feature extraction: an association rule mining based study on cancer expression and methylation profiles

3PNMF-MKL: A non-negative matrix factorization-based multiple kernel learning method for multi-modal data integration and its application to gene signature detection

Optimal ranking and directional signature classification using the integral strategy of multi-objective optimization-based association rule mining of multi-omics data

Contact Info

Product

Resources

About