Figure 11

Theoretical functions of protein size distribution. One of the most notorious differences between the distribution models is how they start from cero. The monotonic decrease function is based on the occurrence of stop codons (3/64) along a random DNA sequence (random ORF). In the monotonic model, an ORF of size x is always more frequent than an ORF of size x + 1. The log normal function starts flat and then it increases sharply and peaks at ~200 aa. The gamma function starts steep, but then it peaks flatter at 200 aa than the other models. The sum of exponential function starts even steeper from cero and simulates somehow a monotonic function with a sharp cutoff of ORFs smaller than 33 aa (randomly expected ORFs). The model parameters to plot the density function in R were: dlnorm(x,5.772,0.729), dgamma(x,2.08,0.005), dsexp(x,0.01895,0.0042).