The problem of estimating the relatedness of a pair of biological sequ
ences is addressed. A stochastic model of sequence evolution is descri
bed that allows insertion and deletion as well as replacement of amino
acid residues (or substitution of nucleotides) over time. An expectat
ion-maximization (EM) algorithm that obtains maximum likelihood estima
tes of the model parameters is introduced. The method assumes that the
sequences are related by descent from a common ancestor but the align
ment (i.e., the precise evolutionary correspondence between residues i
n each sequence) is unknown. Results from the E-step of the EM algorit
hm are used to assess the likelihood that any two residues are related
by direct descent from a common ancestor.