#include <iostream>
#include <vector>
#include <algorithm>
#include <atcoder/modint>
using namespace std;
using i32 = int32_t;
using u32 = uint32_t;
using i64 = int64_t;
using u64 = uint64_t;
#define rep(i,n) for(int i=0; i<(n); i++)

using m32 = atcoder::modint1000000007;

int main(){
  int A,B,X,Y; cin >> A >> B >> X >> Y;
  vector<m32> F,I,iF;
  int Z = A+B+abs(X)+abs(Y);
  F.assign(Z+1,1);
  for(int i=1; i<=Z; i++) F[i] = F[i-1] * i;
  I.assign(Z+1,1);
  for(int i=2; i<=Z; i++) I[i] = -1000000007/i * I[1000000007%i];
  iF.assign(Z+1,1);
  for(int i=1; i<=Z; i++) iF[i] = iF[i-1] * I[i];

  auto Comb = [&](int n,int r)->m32{ if(n<r) return 0; return F[n]*iF[r]*iF[n-r]; };

  int cR = (B+4)/4;
  int cU = (B+3)/4;
  int cL = (B+2)/4;
  int cD = (B+1)/4;
  
  if(X < 0){ swap(cR,cL); X = -X; }
  if(Y < 0){ swap(cU,cD); Y = -Y; }

  int leftA = A - X - Y;
  if(leftA < 0 || leftA % 2 != 0){ cout << 0 << endl; return 0; }

  if(X > 0 && cR == 0){ cout << 0 << endl; return 0; }
  if(Y > 0 && cU == 0){ cout << 0 << endl; return 0; }

  m32 ans = 0;

  for(int i=0; i*2<=leftA; i++){
    int j = leftA/2 - i;
    if(i != 0 && cL == 0) continue;
    if(j != 0 && cD == 0) continue;
    m32 tmp = 1;
    if(i != 0) tmp *= Comb(i+cL-1,i);
    if(j != 0) tmp *= Comb(j+cD-1,j);
    if(j+X != 0) tmp *= Comb(i+X+cR-1,i+X);
    if(j+Y != 0) tmp *= Comb(j+Y+cU-1,j+Y);
    ans += tmp;
  }
  cout << ans.val() << endl;
  return 0;
}




struct ios_do_not_sync {
    ios_do_not_sync() {
        ios::sync_with_stdio(false);
        cin.tie(nullptr);
    }
} ios_do_not_sync_instance;