Let $\mathcal F$ be a sheaf of $\mathcal O_X$ modules on a scheme $X$. Fix an affine open subset $U$. If $M$ is a module over the coordinate ring of $U$, we let $\tilde{M}$ denote the associated sheaf of modules.
Why do we have a morphism of sheaves
$$\mathcal F (U)^\tilde{} \rightarrow \mathcal F|_U?$$
Of course on the set $U$ this is obvious (it's the identity), but I don't see how to define the morphism on subsets $V\subset U$. It suffices to do it just on principal open subsets. On these, we know that $ F (U)^\tilde{}$ has a nice description: on $D(f)$, it is $F (U)^\tilde{}_f$. But I don't know what $\mathcal F_U$ is on such sets, so I don't see how to define the homomorphism.
(Perhaps I am missing something obvious -- it is quite late...)
A reference to this morphism appears on page 160 of Liu.
Note that it suffices to take $X = \mathrm{Spec}(A)$ and to construct a homomorphism $$ (\Gamma(X, \mathscr{F}))^\sim \longrightarrow \mathscr{F}. $$ On an open subset $U = D(f)$ $(f \in A)$, the homomorphism $$ \Gamma(X, \mathscr{F})_f \to \Gamma(D(f), \mathscr{F}) $$ is induced by the restriction homomorphism $\Gamma(X,\mathscr{F}) \to \Gamma(U,\mathscr{F})$. To see this, note that $U$ is precisely the set of points $x \in X$ such that $f_x \in \mathscr{O}_x$ is invertible.