Etting a target FDR threshold of 1 at the peptide level. Mass spectrometric analysis resulted in identification of a total of 20,783 peptides. After removing peptides not labelled with all the four labels (n = 212) and those (n = 1968) shared between multiple proteins, 18,603 peptides were considered for identification of proteins. The labelling efficiency was thus 99 . Relative quantitation of proteins was carried out based on the intensities of reporter ions released during MS/ MS fragmentation of peptides. The average relative intensities of the two reporter ions for each of the unique peptide MK-1439 site identifiers for a protein were used to determine relative quantity of a protein and percentage variability. Appropriate filters at the level of peptides/peptide spectral matches (PSMs) and then at the protein level were applied to the quantification values as described in earlier publication20. In brief, Only PSMs that are `unique’ for a protein were included for fold change calculation. Next, PSMs with more than 30 co-efficient of variation ( CV) between the replicate label measurements (i.e., 114 and 115 for control) and (i.e., 116 and 117 for tumor) were removed programmatically. We then extracted PSMs corresponding to proteins with 1.5 fold change, applied 1.5 fold cut off to these subset of PSMs and recomputed fold change for proteins. Further filters were applied at protein level to select proteins with minimum 2 unique peptides and 2-fold expression change, with PSM quant ratio Decumbin custom synthesis variability ( CV) of less than 40 . The median pair-wise quant ratio for 116/114, 116/115, 117/114, and 117/115 was used to compute the statistical significance (p-value < 0.05). The Benjamini Hochberg FDR corrected p-value is included in Supplementary Table S1 for proteins that were differential at 2-fold-change or above. Gene Ontology annotations of the proteins identified were carried out based on Human Protein Reference Database (HPRD, http://www.hprd.org)21. Mapping of molecular functions and pathways was done using the Ingenuity Pathway Knowledge Base (Ingenuity Systems, Redwood City, CA) tool. Proteins containing signal peptide and transmembrane domains were identified using SignalP 4.1 and TMHMM 2.0 software tools. Exocarta database was used to map the human exosomal proteins22.Bioinformatics analysis.Immunohistochemistry (IHC).The expression level of four of the select proteins, epidermal growth factor receptor (EGFR), brevican core protein (BCAN), ectonucleotide pyrophosphatase/phosphodiesterase family member 6 (ENPP6) and heterogeneous nuclear ribonucleoprotein (HNRNP) K were studied by immunohistochemistry using commercially available Tissue microarray containing 13 Diffuse Astrocytomas cases and 4 control tissue cores (US BioMax). In brief, after deparaffinization and rehydration of formalin-fixed paraffin-embedded tumor tissue sections, antigen retrieval was performed by immersing the slide in antigen retrieval buffer (10 mM sodium citrate, 0.05 Tween 20, pH 6.0) at 95 for 5 min. Endogenous peroxidases were blocked with 0.03 hydrogen peroxide, and nonspecific binding was blocked with 2 fetal calf serum in Tris-buffered saline with 0.1 Triton X-100 (TBST, pH 7.6). Sections were then incubated for 1 h at RT with EGFR (dilution 1:100; Cat No. HPA018530), BCAN (dilution-1:200; Cat No. HPA007865), ENPP6 (dilution-1:10; Cat No. HPA042740) and HNRNP K (dilution-1:250; Cat No. HPA007644) primary antibodies (Atlas Antibodies, Sigma) followed by pero.Etting a target FDR threshold of 1 at the peptide level. Mass spectrometric analysis resulted in identification of a total of 20,783 peptides. After removing peptides not labelled with all the four labels (n = 212) and those (n = 1968) shared between multiple proteins, 18,603 peptides were considered for identification of proteins. The labelling efficiency was thus 99 . Relative quantitation of proteins was carried out based on the intensities of reporter ions released during MS/ MS fragmentation of peptides. The average relative intensities of the two reporter ions for each of the unique peptide identifiers for a protein were used to determine relative quantity of a protein and percentage variability. Appropriate filters at the level of peptides/peptide spectral matches (PSMs) and then at the protein level were applied to the quantification values as described in earlier publication20. In brief, Only PSMs that are `unique' for a protein were included for fold change calculation. Next, PSMs with more than 30 co-efficient of variation ( CV) between the replicate label measurements (i.e., 114 and 115 for control) and (i.e., 116 and 117 for tumor) were removed programmatically. We then extracted PSMs corresponding to proteins with 1.5 fold change, applied 1.5 fold cut off to these subset of PSMs and recomputed fold change for proteins. Further filters were applied at protein level to select proteins with minimum 2 unique peptides and 2-fold expression change, with PSM quant ratio variability ( CV) of less than 40 . The median pair-wise quant ratio for 116/114, 116/115, 117/114, and 117/115 was used to compute the statistical significance (p-value < 0.05). The Benjamini Hochberg FDR corrected p-value is included in Supplementary Table S1 for proteins that were differential at 2-fold-change or above. Gene Ontology annotations of the proteins identified were carried out based on Human Protein Reference Database (HPRD, http://www.hprd.org)21. Mapping of molecular functions and pathways was done using the Ingenuity Pathway Knowledge Base (Ingenuity Systems, Redwood City, CA) tool. Proteins containing signal peptide and transmembrane domains were identified using SignalP 4.1 and TMHMM 2.0 software tools. Exocarta database was used to map the human exosomal proteins22.Bioinformatics analysis.Immunohistochemistry (IHC).The expression level of four of the select proteins, epidermal growth factor receptor (EGFR), brevican core protein (BCAN), ectonucleotide pyrophosphatase/phosphodiesterase family member 6 (ENPP6) and heterogeneous nuclear ribonucleoprotein (HNRNP) K were studied by immunohistochemistry using commercially available Tissue microarray containing 13 Diffuse Astrocytomas cases and 4 control tissue cores (US BioMax). In brief, after deparaffinization and rehydration of formalin-fixed paraffin-embedded tumor tissue sections, antigen retrieval was performed by immersing the slide in antigen retrieval buffer (10 mM sodium citrate, 0.05 Tween 20, pH 6.0) at 95 for 5 min. Endogenous peroxidases were blocked with 0.03 hydrogen peroxide, and nonspecific binding was blocked with 2 fetal calf serum in Tris-buffered saline with 0.1 Triton X-100 (TBST, pH 7.6). Sections were then incubated for 1 h at RT with EGFR (dilution 1:100; Cat No. HPA018530), BCAN (dilution-1:200; Cat No. HPA007865), ENPP6 (dilution-1:10; Cat No. HPA042740) and HNRNP K (dilution-1:250; Cat No. HPA007644) primary antibodies (Atlas Antibodies, Sigma) followed by pero.