//Proud of You//
#include <bits/stdc++.h>
#pragma GCC optimization ("O3")
#pragma GCC optimization ("unroll-loops")
using namespace std;
const int N = 1e6 + 17, M = 1e9 + 7;
int n, m, k, id[N], mx[N], c[N], f[N];
vector <int> adj[N], topo;
queue <int> q;
int main()
{
ios_base::sync_with_stdio(0);
cin.tie(0);
cout.tie(0);
cin >> n >> m >> k;
for (int i = 1, u; i <= k; ++i)
{
cin >> u;
id[u] = i;
}
for (int i = 1, u, v; i <= m; ++i)
{
cin >> u >> v;
adj[u].push_back(v);
++c[v];
}
for (int u = 1; u <= n; ++u)
{
if (!c[u])
{
q.push(u);
}
}
while (!q.empty())
{
int u = q.front();
q.pop();
topo.push_back(u);
mx[u] = max (mx[u], id[u]);
for (int v: adj[u])
{
--c[v];
mx[v] = max (mx[v], mx[u]);
if (!c[v])
{
q.push(v);
}
}
}
f[1] = 1;
for (int u: topo)
{
for (int v: adj[u])
{
if (mx[v] == mx[u] || id[v] == mx[u] + 1)
{
f[v] = (f[v] + f[u]) % M;
}
}
}
cout << f[n];
}
