Figure 2a and b show all Boolean functions with one and two inputs, respectively. Each Boolean function is represented Temsirolimus structure by a truth table where for each imput the output 0 or 1 is specified. The letters A and B are used to denote the inputs and the b index of each function is indicated on the upper raw of the truth table. We note that functions where the output is independent of at least one input are not considered, because they can be reduced to a simpler function. For example func tion is equivalent to have no markers assigned and function is equivalent to after removing the marker B. To explore different Boolean functions we change the function, add a new marker or remove one marker. When changing a Boolean function, ��, a new function is selected at random among all consid ered Boolean functions with the same number of in puts.
When removing a marker, ��, if the drug has one marker then we remove it, the drug will have no markers assigned and, therefore, it will not be considered for the treatment of any patient. If the drug has two markers assigned then we remove one of the two markers and use the transformations illustrated in Figure 2c and d. For example, in Figure 2c we start with the function and remove the B input. For this function the output is always 0 when the A input is 1 but the output can be 0 or 1 when the A input is 0. Therefore, can be mapped to or after removing the B input. Since the output of is independent of the input state it is not consid ered. A similar reasoning can be applied to obtain the mappings for function when removing the A marker instead.
Applying this approach to every function we obtain the mappings in Figure 2e and f. Fi nally, if a marker is added, ��, then we use the mappings in Figure 2g, which are the reverse of �� removing the A input. In all cases, when more that one choice is available we choose Anacetrapib one of them with equal probability. Case study To test our methodology we investigate an in silico case study where we can actually quantify the response of each sample to each drug. The in silico case study is based on in vitro growth inhibition data reported by the Sanger Institute. In the Sanger screen 714 cell lines were tested for their responses against 138 drugs. For several sample drug pairs the natural logarithm of the drug concentration to achieve a 50% growth inhibition relative to untreated controls was reported.
The logIC50 data is missing for 26,031 drug cell line pairs, representing 20% of all drug sample pairs. The missing logIC50 data was imputed using the weighted http://www.selleckchem.com/products/Tipifarnib(R115777).html average approach described in the Methods section. The Pearson Correlation Coefficient between the im puted and actual log50s, when the latter were available, was 0. 89. For each cell line the cancer subtype and the status of 47 cancer related genes was also reported, including somatic mutations and copy number alterations.