The integrated analysis of metabolic and protein interaction networks reveals novel molecular organizing principles
Bioinformatics Group, Max Planck Institute for Molecular Plant Physiology, Am Mühlenberg 1, 14424 Potsdam-Golm, Germany
BMC Systems Biology 2008, 2:100 doi:10.1186/1752-0509-2-100Published: 25 November 2008
Additional file 1:
Raw counts, pairwise network (PIN, MIN) distances for Figure3. Enrichment and depletion of the abundance of shortest path pairs between nodes represented in both the PIN and MIN. The enrichments and depletions were judged by the z-score (number in cell) of the frequency of observations (number in paranthesis) in comparison to randomized distribution with red-color indicating enrichment and blue-color depletion relative to randomized networks. Distances > 14 comprise distances greater than 14 and not connected distance pairs. Blank cells indicates 0 path pairs with z-Score of 0.
Format: XLS Size: 64KB Download file
This file can be viewed with: Microsoft Excel Viewer
Additional file 2:
Dependency of PIN-MIN shortest path correlations as a function of considered maximal distance. Pearson correlation coefficient of PIN and MIN distance pairs as a function of considered maximally allowed shortest path distances in PINs and MINs. The correlation of distance pairs were calculated only including distance pairs shorter than the indicated cutoff distances dPIN and dMIN. A) rPIN and EIN; B) fPIN and EIN; C) ePIN and EIN; D) rPIN and mapEIN; E) fPIN and mapEIN; F) ePIN and mapEIN; G) rPIN and CIN; H) fPIN and CIN and I) ePIN and EIN I). Generally, a trend towards more pronounced correlations with decreasing cutoff distance is evident. Note: The drop in correlation values for short distance cutoff values is primarily explained by the inevitable loss of correlation when the considered absolute range is reduced. This is purely a statistical effect. For any correlated, but scattered data, correlation coefficients inevitably drop, if the considered range is reduced.
Format: JPEG Size: 171KB Download file
Additional file 3:
Detected physical interaction of enzymes involved in selected pathways. Detected physical interaction of enzymes involved in selected pathways. A) ergosterol biosynthesis; B) sphingolipid biosynthesis; C) ubiquinone biosynthesis; D) fatty acid biosynthesis; E) glycogen biosynthesis; F) last step of polyamine biosynthesis; G) de novo biosynthesis of pyrimidine ribonucleotides; H) superpathway of leucine, isoleucine, and valine biosynthesis; I) lysine biosynthesis; J) superpathway of phenylalanine, tyrosine and tryptophan biosynthesis. In picture A) HMG1/2, MVD1, IDI1 and ERG10/13 are part of the mevalonate pathway. All pathways are derived from the SGD Database. Only enzymes contained in the PIN are visualized, i.e. the pathways are not complete in a biochemical sense. For the fatty acids biosynthesis, the malic enzymes (MAE) as well as the malate dehydrogenase (MDH2) were included as sources of NADPH and AcetylCoA. Enzymes are abbreviated by their gene symbols and detected interactions between them are denoted by connecting lines.
Format: JPEG Size: 123KB Download file
Additional file 4:
Supplementary Annotation and Raw Count Data. GO categories used for creating filtered version of protein interaction networks; List of all 51 molecules (currency metabolites, co-factors) excluded from the analysis.
Format: PDF Size: 40KB Download file
This file can be viewed with: Adobe Acrobat Reader