speed up the & operation

How to speed up the following code? I did profiling, and this line takes 3267.72s for calling 245913489 times.

a=[1 0 1 1 0 1 1 1 1 1 0 0 1 1 0 1 1 1 0 1 1 0 1 1 1 0 1 1 0 1 1 1];
b=[1 0 1 1 0 1 1 1 1 1 0 0 1 1 0 1 1 1 0 1 1 0 1 1 1 0 1 1 0 1 1 1  1 1 1 1 1 0 0 1 1 0 1 1 1 0 1 1 0 1 1 1 0];
for i=1:1000000
aat=((a(1,1:24)==0)&(b(1,25:48)==1)&(~ismembc(1:24,5:50)))';
end

[dpb]

See other comments on addressing being overhead-intensive.

I'd also guess it's highly likely the time is in ismembc as compared to &

[Adam]

Break the line up into all its components and profile again.

[Chaoyang Jiang]

Could you please explain it more explicitly? Thank you.

[Chaoyang Jiang]

ismembc seems to be faster than ismember, that is why I used this function.

[Walter Roberson]

for i=1:1000000
  part1 = a(1,1:24)==0;
  part2 = b(1,25:48)==1;
  part3 = ~ismembc(1:24,5:50);
  part4 = part1 & part2 & part3;
  aat = part4 .';
end

Now when you profile you can see exactly which part of the expression is taking the most time.

We suspect that it is the ismembc() that is taking all of the time and that the & is relatively fast.