Use of fragment-sharing estimates from DNA fingerprinting to determine relatedness in a tropical wren

We assessed the usefulness of DNA fragment-sharing scores from DNA fingerprints for assigning relatedness to unknown pairs of individuals in a population of stripe-backed wrens (Campylorhynchus nuchalis). Preliminary investigation of scoring biases revealed consistency both within and between scorers in relative band-sharing scores, but a tendency for scores to be inflated and for inter-scorer agreement to decline as distance between lanes on an autoradiograph increased. Distributions of band-sharing values matched expected distributions well, which suggests that variability in scores is mostly inherent and not a result of errors in scoring. Confidence intervals based on band-sharing scores or means of scores across enzymes, probes and scorers revealed that unrelated (r = 0) and first-order dyads (r = 1/2) could be distinguished on the basis of single band-sharing scores from the best combination of enzyme and probe (HaeIII/33.15) and that first- and second-order dyads could be distinguished when confidence intervals were based on means of band-sharing scores across two enzymes, two probes and two scorers.