Sequence length and error analysis of Sequenase and automated Taq cycle sequencing methods.

We have examined DNA sequence error as a function of length using both a manual method of performing reactions with Sequenase and an automated Taq cycle sequencing method. DNA fragments from both methods were separated and analyzed on a sequencer. To determine the sequence of a cosmid insert (35.3 kb), 379 sequences were obtained from a manual Sequenase method, and 354 sequences were obtained from a Taq cycle sequencing method as performed on an automated robotic workstation and sequenced on an automated fluorescent sequencer. A highly redundant consensus of these sequences was obtained and aligned with the individual sequences to determine sequence error over the length of each sequence. The results of this study indicate that error is about 1% per position over the first 350 nucleotides, but increases thereafter to about 17% at 500 nucleotides. This pattern of accuracy was nearly equivalent for manual Sequenase methods and automated Taq cycle sequencing methods. The potential of these methods in large-scale DNA sequencing projects is discussed.