Multiplex DNA Sequencing 



Robert Weiss and Raymond F. Gesteland 



Howard Hughes Medical Institute, Department of Human Genetics, University of Utaii 



Medical School, Salt Lake City, UT 841 12 



(801)581-5190 



We have developed a method for rapid DNA sequencing that employs multiplex 

 probing, as originally described by Church and Gilbert. A large number (25-50) of 

 DNA samples, each cloned in an equal number of special vectors, are prepared together 

 and as a mixture are sequenced using dideoxy sequencing from a universal primer 

 whose sequence is carried by each vector. After conventional separation by gel 

 electrophoresis, the mixed DNA pattern is transferred to a membrane. Each of the 

 individual sequences is then revealed by repetitive rounds of probing, washing, reading, 

 stripping, and reprobing with labeled oligonucleotide, each of which is unique for a 

 sequence in one of the vectors. By fixing a number of such membranes in a drum, the 

 '-P-labeled probing can be done (without handling the membranes) to obtain 

 10.000-20,000 bases of sequence in each cycle with a cycle time of 1-2 days, including 

 time for autoradiography — with minimal labor. By using a set of transposons containing 

 appropriate primer and identifier sequences, we are developing, with help from Diane 

 Dunn, a method for creating in vivo random subclones that are ready for multiplex 

 sequencing. An optical lab has been set up by Jeff Ives and Achim Karger with the help 

 of Joel Harris (Chemistry Department) to compare the feasibility of fluorescence and 

 chemiluminescent tags for the multiple probes. The efficiency of multiplex sequencing 

 has been diminished by a bottleneck at the step of reading autoradiograms. To solve this 

 problem and to deal with data that might come from charge-coupled display (CCD) 

 images of the fluorescent or chemiluminescent membranes, a research group — 

 including Jeff Ives, Mike Murdock, and Tom Stockham and Neil Cotter (Electrical 

 Engineering Department) — has been assembled. Harold Swerdlow is investigating the 

 feasibility of using gel electrophoresis in microbore capillaries (70 mm) to do 

 sequencing. 



91 



