Figure 4

Analysis of the T cell network. (a) The T cell network [24]. The channel input molecules are TCR lig, CD4 and CD28, whereas the output molecules are AP1, bcat, BclXL, CRE, Cyc1, FKHR, NFκB, p21c, p27k, p38, p70S6K, SRE, NFAT and SHP2. Green arrows represent activatory interactions and red blunt lines show inhibitory interactions. This figure is intended to provide a general picture of the network. For specific details and regulatory mechanisms of each molecule, one can refer to the equations listed in Additional file 1: Table S1. (b) Values of transmission error probability P _e and capacity C are calculated for different molecules in the network with the output node SHP2, as an example output molecule. P _e and C values for those molecules not listed in this table are calculated as 0 and 1, respectively. See Additional file 1 for the list of these molecules. (c) Transmission error probability P _e versus the dominance factor k in the T cell communication channel for the two molecules PI3K and Fyn. When PI3K is the dominant dysfunctional molecule, P _e rapidly increases as the dominance factor k increases. This is in agreement with experimental data, which shows when PI3K is knocked out (inhibited with both Ly294002 and Wortmannin), PKB does not properly receive signals from the input molecules and remains inactive (its phosphorylation is blocked in human T cells) [24]. In contrast, the decrease of P _e when Fyn is the dominant dysfunctional molecule means that even if Fyn is dysfunctional with probability one, there will be no transmission error. This is consistent with the experimental observation that when Fyn is knocked out (Fyn-deficient and heterozygous splenic mouse T cells), stimulation of the input molecules still correctly regulates PKB [24].