Connecting the dots: The boons and banes of network modeling

Abstract

Summary Network modeling transforms data into a structure of nodes and edges such that edges represent relationships between pairs of objects, then extracts clusters of densely connected nodes in order to capture high-dimensional relationships hidden in the data. This efficient and flexible strategy holds potential for unveiling complex patterns concealed within massive datasets, but standard implementations overlook several key issues that can undermine research efforts. These issues range from data imputation and discretization to correlation metrics, clustering methods, and validation of results. Here, we enumerate these pitfalls and provide practical strategies for alleviating their negative effects. These guidelines increase prospects for future research endeavors as they reduce type I and type II (false-positive and false-negative) errors and are generally applicable for network modeling applications across diverse domains.

RWRtoolkit: multi-omic network analysis using random walks on multiplex networks in any species

“…Biological studies are increasingly pursuing and obtaining data on larger scales and at multiple levels in the molecular hierarchy of the study system. One approach for dealing with the multiplicity of data in modern biology is to represent the relationships in the data as a network [1]. Each entity in a dataset (e.g., each gene) becomes a node, and an edge between two nodes represents a relationship that has been measured or predicted between those nodes (e.g., their co-expression in a population, sharing of common protein domains, similarity of methylation state, etc.).…”

Heterogeneity impacts biomarker discovery for precision medicine

“…We previously reported that the use of standard correlation metrics in network modeling leads to increased type II errors in the presence of subtype groups 5,7,9,11 . All the correlation measures that we have examined, including Pearson’s correlation coefficient 12 , r -squared 13 , dot product 14 , and mutual information 15 , return single scalar values that are crippled by heterogeneity 5 . They are universal measures, in that individuals in the entire group are viewed as a whole and thus subtle but crucial subgroup structures, which manifest heterogeneity of the individuals in the group, are ignored.…”

“…They are universal measures, in that individuals in the entire group are viewed as a whole and thus subtle but crucial subgroup structures, which manifest heterogeneity of the individuals in the group, are ignored. If two analytes are highly correlated for a subset of individuals but not at all correlated for the others, the correlation value is reduced due to the latter individuals, thereby contributing to false negative signals 5,11 .…”

“…Popular statistics for this domain include fold change (FC) of levels of candidate biomarkers between diseased cases and normal controls and area under the receiver operating characteristic curve (AUC) 3 . In addition to tests on individual biomarkers, patterns comprised of multiple biomarkers can be identified by modeling the data as a network in which biomarkers are represented by nodes and correlations between pairs of biomarkers are represented by edges between the corresponding node pairs [4][5][6][7][8][9] . Clusters of intercorrelated biomarkers are identified and evaluated using methods such as modularity 7,10 or gene enrichment 11 .…”

“…Another major challenge is that traditional statistical methods that are successful for global biomarkers can be inappropriate for subset biomarker identification, and each approach needs to be reevaluated for use in this distinct domain. We previously reported that the use of standard correlation metrics in network modeling leads to increased type II errors in the presence of subtype groups 8,9,12,13 . All the correlation measures that we have examined, including Pearson’s correlation coefficient 14 , r -squared 15 , dot product 16 , and mutual information 17 , return single scalar values that are crippled by heterogeneity 9 .…”