Joint analysis of multiple high-dimensional data types using sparse matrix approximations of rank-1 with applications to ovarian and liver cancer

Abstract

BackgroundTechnological advances enable the cost-effective acquisition of Multi-Modal Data Sets (MMDS) composed of measurements for multiple, high-dimensional data types obtained from a common set of bio-samples. The joint analysis of the data matrices associated with the different data types of a MMDS should provide a more focused view of the biology underlying complex diseases such as cancer that would not be apparent from the analysis of a single data type alone. As multi-modal data rapidly accumulate in research laboratories and public databases such as The Cancer Genome Atlas (TCGA), the translation of such data into clinically actionable knowledge has been slowed by the lack of computational tools capable of analyzing MMDSs. Here, we describe the Joint Analysis of Many Matrices by ITeration (JAMMIT) algorithm that jointly analyzes the data matrices of a MMDS using sparse matrix approximations of rank-1.MethodsThe JAMMIT algorithm jointly approximates an arbitrary number of data matrices by rank-1 outer-products composed of “sparse” left-singular vectors (eigen-arrays) that are unique to each matrix and a right-singular vector (eigen-signal) that is common to all the matrices. The non-zero coefficients of the eigen-arrays identify small subsets of variables for each data type (i.e., signatures) that in aggregate, or individually, best explain a dominant eigen-signal defined on the columns of the data matrices. The approximation is specified by a single “sparsity” parameter that is selected based on false discovery rate estimated by permutation testing. Multiple signals of interest in a given MDDS are sequentially detected and modeled by iterating JAMMIT on “residual” data matrices that result from a given sparse approximation.ResultsWe show that JAMMIT outperforms other joint analysis algorithms in the detection of multiple signatures embedded in simulated MDDS. On real multimodal data for ovarian and liver cancer we show that JAMMIT identified multi-modal signatures that were clinically informative and enriched for cancer-related biology.ConclusionsSparse matrix approximations of rank-1 provide a simple yet effective means of jointly reducing multiple, big data types to a small subset of variables that characterize important clinical and/or biological attributes of the bio-samples from which the data were acquired.Electronic supplementary materialThe online version of this article (doi:10.1186/s13040-016-0103-7) contains supplementary material, which is available to authorized users.

A neural network model to classify liver cancer patients using data expansion and compression

“…We consider two sets of genes, one including 36 genes and another one including 40 genes. These genes have been found by analyzing larger genomic data sets by the authors, as described in [7]. The genes in this signatures carry signal information that classify the ovarian cancer patients by their response to standard chemotherapy.…”

“…The high number of parameters increases the calculation complexity for the classification of the data in the neural networks [1]- [4]. The authors have developed an algorithm to reduce the number of genes (parameters) to less than 40 [5]- [7]. We consider the data as a matrix, in which rows are the genes and columns are samples (patients).…”

Capecitabine inhibits epithelial‑to‑mesenchymal transition and proliferation of colorectal cancer cells by mediating the RANK/RANKL pathway

“…RANK/RANKL can also be exploited as a target for the treatment of myeloma and solid tumors, such as prostate and breast cancer ( 22 ). In tissues from liver ( 23 ), stomach ( 24 ), breast ( 25 ), thyroid ( 26 ), prostate ( 27 ) and pancreatic cancers ( 28 ), high levels RANK expression were previously detected, suggesting that RANK may be involved in the occurrence and development of malignant tumors. Therefore, RANK may apply a multitude of mechanisms to facilitate the tumorigenic process.…”

JAMMIT Analysis Defines 2 Semi-Independent Immune Processes Common to 29 Solid Tumors

“…Also, the role of the cells of the innate immune system has seldom been included in analyses of tumor immunophenotype, impeding development of a broad consensus with regard to the role of tumor-associated macrophages (TAM’s) and “M1” and “M2” macrophages in cancer. Members of our group previously developed JAMMIT (Joint Analysis of Multiple Matrices by Iteration 16 ) and applied the technique to both experimentally-acquired data and the publicly-available TCGA database. Here we show results based on JAMMIT analysis of multiple tumor types from TCGA.…”