Extracting proteins involved in disease progression using temporally connected networks.
Autor: | Anand R; Drug Discovery Research Centre, Translational Health Science and Technology Institute, NCR Biotech science cluster, 3rd milestone, Faridabad-Gurgaon Expressway, Faridabad, 121001, India., Sarmah DT; Drug Discovery Research Centre, Translational Health Science and Technology Institute, NCR Biotech science cluster, 3rd milestone, Faridabad-Gurgaon Expressway, Faridabad, 121001, India., Chatterjee S; Drug Discovery Research Centre, Translational Health Science and Technology Institute, NCR Biotech science cluster, 3rd milestone, Faridabad-Gurgaon Expressway, Faridabad, 121001, India. samrat.chatterjee@thsti.res.in. |
---|---|
Jazyk: | angličtina |
Zdroj: | BMC systems biology [BMC Syst Biol] 2018 Jul 25; Vol. 12 (1), pp. 78. Date of Electronic Publication: 2018 Jul 25. |
DOI: | 10.1186/s12918-018-0600-z |
Abstrakt: | Background: Metabolic disorders such as obesity and diabetes are diseases which develop gradually over time in an individual and through the perturbations of genes. Systematic experiments tracking disease progression at gene level are usually conducted giving a temporal microarray data. There is a need for developing methods to analyze such complex data and extract important proteins which could be involved in temporal progression of the data and hence progression of the disease. Results: In the present study, we have considered a temporal microarray data from an experiment conducted to study development of obesity and diabetes in mice. We have used this data along with an available Protein-Protein Interaction network to find a network of interactions between proteins which reproduces the next time point data from previous time point data. We show that the resulting network can be mined to identify critical nodes involved in the temporal progression of perturbations. We further show that published algorithms can be applied on such connected network to mine important proteins and show an overlap between outputs from published and our algorithms. The importance of set of proteins identified was supported by literature as well as was further validated by comparing them with the positive genes dataset from OMIM database which shows significant overlap. Conclusions: The critical proteins identified from algorithms can be hypothesized to play important role in temporal progression of the data. |
Databáze: | MEDLINE |
Externí odkaz: |