A method for determining the sequence of at least a portion of a single-stranded nucleic acid molecule by base-specifically labeling the exposed bases of the nucleic acid molecule using heavy element-substituted nucleotide bases which form Watson-Crick type base-pairs with the exposed bases of the nucleic acid molecule and then imaging the labeled single-stranded nucleic acid molecule using electron microscopy, e.g., transmission electron microscopy (TEM), or some other method that permits discrimination of the heavy element substituted nucleotide bases is described. The image is analyzed to determine the base sequence of at least a portion of the nucleic acid molecule.