Skip to main content
Figure 5 | BioData Mining

Figure 5

From: Graph representation of high-dimensional alpha-helical membrane protein data

Figure 5

The Workflow. The workflow for membrane environment information extraction and transformation. A: For each membrane protein, all possible membrane helices have been predicted using TMHMM. Predicted ‘TM’ sequence information is coloured in red and ‘nTM’ in blue. B: After deriving ‘TM’ sequence information all possible motifs with n-1 highly variable positions by n = {4 - 7} were determined by using a common naive text search algorithm (Figure 2). Further, for each ‘TM’ sequence part, all possible MAs consisting of four directly consecutive motifs have been detected. C: The later applying of useful and powerful algorithms which are involved in the statistical information aggregation assumes, that each detected (MA x , TM y ) is considered to be a graph structure. This leads to the transfer of each (MA x , TM y ) into a graph where each motif can be considered as a node connected by a edge to the following node. D: Finally, all ‘TM’ sequence part corresponding graphs were merged into one. The edge-weightiness of the already existing source and target nodes were updated by increasing by one. Ultimately, a weighted graph exists for each ‘TM’ sequence part which leads to the final merge process and the resulting graph.

Back to article page