Revealing Clusters of Connected Pathways Through Multisource Data Integration in Huntington's Disease and Spastic Ataxia

Abstract

The advancement of scientific and medical research over the past years has generated a wealth of experimental data from multiple technologies, including genomics, transcriptomics, proteomics and other forms of -omics data, which are available for a number of diseases. The integration of such multisource data is a key component towards the success of precision medicine. In this work we are investigating a multisource data integration method developed by our group, regarding its ability to drive to clusters of connected pathways under two different approaches: (a) a disease-centric approach, where we want to integrate data around a disease, and (b) a gene-centric approach, where we want to integrate data around a gene. We have used as a paradigm for the first approach Huntington's disease (HD), a disease with a plethora of available data, whereas for the second approach the GBA2, a gene that is related to Spastic Ataxia (SA), a phenotype with sparse availability of data. Our work shows that valuable information at the level of disease-related pathway clusters can be obtained for both HD and SA. New pathways that classical pathway analysis methods were unable to reveal, are emerging as necessary "connectors" to build connected pathway stories formed as pathway clusters.

Inhibition of Breast Cancer Cell Invasion by Ras Suppressor-1 (RSU-1) Silencing Is Reversed by Growth Differentiation Factor-15 (GDF-15)

“…For the detection of disease-related clusters of molecular mechanisms, we used a recently introduced methodology utilizing a reference pathway network [80,81,82]. The specific reference network is based on the functional connection among pathways that can be found in the Kyoto Encyclopedia of Genes and Genomes (KEGG) repository (Available online: , last accessed in March 2018), which is a well-established database for obtaining pathway information.…”

“…Specifically, for a given set of pathways, this methodology identifies and adds the key pathways (nodes) that ensure the minimal connectivity of the network. For this, a specific algorithm calculates all the shortest paths within the reference network that interconnect these nodes together [80,81,82]. This process results in an extended network of pathways able to provide enriched functional information related to the phenotype under study.…”

Investigating the Transition of Pre-Symptomatic to Symptomatic Huntington’s Disease Status Based on Omics Data

“…The HD related pathways which were identified in the pre-symptomatic and symptomatic HD patients using PathwayConnector [ 15 ] were then compared to the HD-related pathways obtained through our multi-source data integration approach [for more details on methods and findings see 33]. Some of the common identified pathways between pre-symptomatic HD and symptomatic HD patients and HD related pathways were identified from the multi-source data integration approach [ 41 ]. Some of the common pathways include, cocaine addiction, dopaminergic synapse, pathways in cancer, and several other pathways.…”

ProtExA: A tool for post-processing proteomics data providing differential expression metrics, co-expression networks and functional analytics

“…In classical pathway analysis gene lists, usually obtained from any experimental-computational method, can be further analysed by relevant software tools that allow enrichment analysis to be performed based on prior knowledge gene-set libraries and pathways connected to them. Such tools may provide significant score-based information on how genes are involved into pathways and in effect, how pathways to a disease [28] . In this line, ProTExA can further perform enrichment analysis to the list of top-scored protein-genes that eventually derive from statistical analysis and filtering through the EnrichR package [13] , which has been widely used to identify involved pathways from specific lists of genes.…”