Discovering functional relationships between RNA expression and chemotherapeutic susceptibility using relevance networks

Abstract

In an effort to find gene regulatory networks and clusters of genes that affect cancer susceptibility to anticancer agents, we joined a database with baseline expression levels of 7,245 genes measured by using microarrays in 60 cancer cell lines, to a database with the amounts of 5,084 anticancer agents needed to inhibit growth of those same cell lines. Comprehensive pair-wise correlations were calculated between gene expression and measures of agent susceptibility. Associations weaker than a threshold strength were removed, leaving networks of highly correlated genes and agents called relevance networks. Hypotheses for potential single-gene determinants of anticancer agent susceptibility were constructed. The effect of random chance in the large number of calculations performed was empirically determined by repeated random permutation testing; only associations stronger than those seen in multiply permuted data were used in clustering. We discuss the advantages of this methodology over alternative approaches, such as phylogenetic-type tree clustering and self-organizing maps.

Linking pathway gene expressions to the growth inhibition response from the National Cancer Institute's anticancer screen and drug mechanism of action

“…[2][3][4][5][6][7][8][9][10][11][12][13][14][15][16] Previous efforts using this strategy have focused mainly on finding causes of drug resistance. [2][3][4]11,17,18 Gene expression signatures have also been used as surrogate markers of cellular states, for example, to identify agents that induce the differentiation of acute myeloid leukemia cells. 19 However, nearly all of these investigations have been based on single gene expression-drug response relationships, whereas complex interactions between a drug and highly interconnected biological networks may not be reflected solely by the state of any one gene.…”

Reconstruction of Biological Networks by Supervised Machine Learning Approaches

“…In [14] this benchmark was used to compare different algorithms, including Bayesian networks [15], ARACNe [26], and the context likelihood of relatedness (CLR) algorithm [14], a new method that extends the relevance networks class of algorithms [9]. They observed that CLR outperformed all other methods in prediction accuracy, and experimentally validated some predictions.…”

“…Another rationale for de novo inference is to connect genes or proteins that are similar to each other in some sense [25,30], For example, co-expression networks, or the detection of similar phylogenetic profiles are popular ways to infer "functional relationships" between proteins, although the meaning of the resulting edges has no clear biological justification [36]. Similarly, some authors have attempted to predict gene regulatory networks by detecting large mutual information between expression levels of a TF and the genes it regulates [9,14].…”

The future of dermatopathology

“…44 A second use for the oligonucleotide microarray is that of biopharmacology, where anticancer agent efficacy can be correlated to gene expression in tissue. 45,46 Complimentary to these novel genomic initiatives is proteomics, the study of expressed proteins in tissue and cell types. 47 Both proteomic and other tissue microarray studies are enhanced by the use of laser capture microdissection which enables isolation of critical neoplastic cells.…”