Investigation of multiple sclerosis-related pathways through the integration of genomic and proteomic data

Abstract

Background: Multiple sclerosis (MS) has a complex pathophysiology, variable clinical presentation, and unpredictable prognosis; understanding the underlying mechanisms requires combinatorial approaches that warrant the integration of diverse molecular omics data.

Methods: Here, we combined genomic and proteomic data of the same individuals among a Turkish MS patient group to search for biologically important networks. We previously identified differentially-expressed proteins by cerebrospinal fluid proteome analysis of 179 MS patients and 42 non-MS controls. Among this study group, 11 unrelated MS patients and 60 independent, healthy controls were subjected to whole-genome SNP genotyping, and genome-wide associations were assessed. Pathway enrichment analyses of MS-associated SNPs and differentially-expressed proteins were conducted using the functional enrichment tool, PANOGA.

Results: Nine shared pathways were detected between the genomic and proteomic datasets after merging and clustering the enriched pathways. Complement and coagulation cascade was the most significantly associated pathway (hsa04610, P = 6.96 × 10^-30). Other pathways involved in neurological or immunological mechanisms included adherens junctions (hsa04520, P = 6.64 × 10^-25), pathogenic Escherichia coli infection (hsa05130, P = 9.03 × 10^-14), prion diseases (hsa05020, P = 5.13 × 10^-13).

Conclusion: We conclude that integrating multiple datasets of the same patients helps reducing false negative and positive results of genome-wide SNP associations and highlights the most prominent cellular players among the complex pathophysiological mechanisms.

Keywords: Bioinformatics; Genomics; Multiple sclerosis; Proteomics.

Figure 1. Genome-wide Manhattan plot showing –log₁₀ of the P-values of 129.547 SNPs (y-axis) against their genomic positions (x-axis) for associations.

The results are plotted left to right from the p-terminal ends of the chromosomes, which are shown in different colors. The blue line represents the P-value threshold (0.05).

Figure 2. Clustering dendrogram of the enriched pathways.

The dendrogram shows all 151 enriched pathways, which were grouped in clusters based on their relevance to each other. The red line shows the cut-off for biologically relevant clusters, which was cut at the pairwise distance of 0.55, resulting in 33 representative clusters. Thirty-three pathways with the lowest P-values in each cluster constituted the representative pathways.

Figure 3. Complement and coagulation cascades.

The illustration of complement and coagulation cascades shows the identified genes by SNP associations (blue) and differentially expressed proteins (yellow) as a representative pathway for the merged data analysis. The pathway was adapted from KEGG: Kyoto Encyclopedia of Genes and Genomes (Kanehisa & Goto, 2000; hsa04610, 2019) and created with BioRender.com.