notes and speeding up network aging simulation.

The old simulation method is slow due to single_network_failure() 

inpairs$age = rexp( length(inpairs[,1]), rate=lambda )  #exponential ages for pairs
inpairs$age = ifelse(inpairs$active > (1-p), inpairs$age, NA ) #if not active, intxn is excluded. 
  
  ModuleTb = data.frame(runningORFs) #buffer for module ages    
  #loop every essential genes to identify the module age
  for (i in 1:length(runningORFs)) {
    myORF = runningORFs[i]
    pos1 = grep(myORF, inpairs$id1)
    pos2 = grep(myORF, inpairs$id2)  #id1,2 to ORF1,2 is a really bad choice. 
    if( length( c(pos1,pos2))>=1 ) {
      ModuleTb$age.m[i] = max( inpairs$age[c(pos1,pos2)], na.rm=T )   #maximal intxn age -> module age
    } else {
      ModuleTb$age.m[i] = NA; 
    }
  }

Basically, there were too much ifelse, matching, grep(). The grep() probably is the most expensive step.

I propose a new algorithm to calculate by Degree for each essential gene, thereby eliminating the grep() process.