In silico molecular cytogenetics: a bioinformatic approach to prioritization of candidate genes and copy number variations for basic and clinical genome research
Open Access
- 9 December 2014
- journal article
- research article
- Published by Springer Nature in Molecular Cytogenetics
- Vol. 7 (1) , 98
Abstract
The availability of multiple in silico tools for prioritizing genetic variants widens the possibilities for converting genomic data into biological knowledge. However, in molecular cytogenetics, bioinformatic analyses are generally limited to result visualization or database mining for finding similar cytogenetic data. Obviously, the potential of bioinformatics might go beyond these applications. On the other hand, the requirements for performing successful in silico analyses (i.e. deep knowledge of computer science, statistics etc.) can hinder the implementation of bioinformatics in clinical and basic molecular cytogenetic research. Here, we propose a bioinformatic approach to prioritization of genomic variations that is able to solve these problems. Selecting gene expression as an initial criterion, we have proposed a bioinformatic approach combining filtering and ranking prioritization strategies, which includes analyzing metabolome and interactome data on proteins encoded by candidate genes. To finalize the prioritization of genetic variants, genomic, epigenomic, interactomic and metabolomic data fusion has been made. Structural abnormalities and aneuploidy revealed by array CGH and FISH have been evaluated to test the approach through determining genotype-phenotype correlations, which have been found similar to those of previous studies. Additionally, we have been able to prioritize copy number variations (CNV) (i.e. differentiate between benign CNV and CNV with phenotypic outcome). Finally, the approach has been applied to prioritize genetic variants in cases of somatic mosaicism (including tissue-specific mosaicism). In order to provide for an in silico evaluation of molecular cytogenetic data, we have proposed a bioinformatic approach to prioritization of candidate genes and CNV. While having the disadvantage of possible unavailability of gene expression data or lack of expression variability between genes of interest, the approach provides several advantages. These are (i) the versatility due to independence from specific databases/tools or software, (ii) relative algorithm simplicity (possibility to avoid sophisticated computational/statistical methodology) and (iii) applicability to molecular cytogenetic data because of the chromosome-centric nature. In conclusion, the approach is able to become useful for increasing the yield of molecular cytogenetic techniques.Keywords
This publication has 61 references indexed in Scilit:
- Interactome Networks and Human DiseaseCell, 2011
- The DNA Replication Stress Hypothesis of Alzheimer’s DiseaseThe Scientific World Journal, 2011
- Cytoscape 2.8: new features for data integration and network visualizationBioinformatics, 2010
- Disease candidate gene identification and prioritization using protein interaction networksBMC Bioinformatics, 2009
- GIN'n'CIN hypothesis of brain aging: deciphering the role of somatic genetic instabilities and neural aneuploidy during ontogenyMolecular Cytogenetics, 2009
- Partial monosomy 7q34-qter and 21pter-q22.13 due to cryptic unbalanced translocation t(7;21) but not monosomy of the whole chromosome 21: a case report plus review of the literatureMolecular Cytogenetics, 2008
- Walking the Interactome for Prioritization of Candidate Disease GenesThe American Journal of Human Genetics, 2008
- Chromosomal mosaicism goes globalMolecular Cytogenetics, 2008
- Unexplained autism is frequently associated with low-level mosaic aneuploidyJournal of Medical Genetics, 2007
- Gene prioritization through genomic data fusionNature Biotechnology, 2006