Two-Stage Approach for Protein Superfamily Classification

Algorithm 2

NSGA-II( ).

Let 𝑛_𝑖 denotes the domination count i.e the number of solutions

which dominates solution 𝑖.

𝑆_𝑖 denotes set of solutions which solution 𝑖 dominates.

 Initialize 𝑛_𝑖=0 and 𝑆_𝑖=𝜙 for every solution 𝑖∈𝑃.

 for (∀𝑗≠𝑖)  and 𝑗∈𝑝  do

  if  𝑖≼𝑗  then

   Update 𝑆_𝑝=𝑆_𝑝∪𝑗

  else  {𝑗≼𝑖}

   set 𝑛_𝑖=𝑛_𝑖+1.

  end if

  if 𝑛_𝑖=𝑂  then

   𝑃₁=𝑃₁∪(𝑖) where 𝑃₁ denotes the first non-dominated front.

  end if

  set front count 𝐾=1.

 end for

 while 𝑃_𝑘≠𝜙  do

  initialize 𝑄=𝜙 for storing next non-dominated solutions.

  for  ∀𝑖∈𝑃_𝑘  and  ∀𝑗∈𝑆_𝑖  do

   Update 𝑛_𝑗=𝑛_𝑗−1

   if 𝑛_𝑗=0  then

    set 𝑄=𝑄∪𝑗

   end if

  end for

  Set 𝐾=𝐾+1 and 𝑃_𝑘=𝜙

 end while