AT A GLANCE Received his master’s degree in electrical engineering in 1990 from Manhattan College. Formerly designed radar systems for tracking aircraft and missile guidance. Interests include circuit design, software programming, bioengineering, bioinformatics and simulation/modeling, as well as sailing and travel.
Q How did you get into doing microarrays?
AWe started work on our cDNA microarray facility in spring 1997 when there was no commercial microarray equipment available. So we talked to Pat Brown at Stanford and visited Mike Bittnerís facility at the NIH. They were very helpful in showing us what had to be done. We then formulated a plan to design our own equipment and establish a core facility to make microarray chips available to all researchers at Einstein. I was recruited at the time to design and build the cDNA microarray printer and scanner for the facility.
Q How is the microarray facility set up within the university?
AThe cDNA microarray lab is organized as a core facility whereby every lab can purchase microarray slides at nominal cost, perform the experiment, and return the slides back to us for scanning. We have also set up a parallel operation using the Affymetrix GeneChip equipment so users have both options open to them.
Q What types of microarrays do you use and in what combination?
AWe prepare our own PCR products from several libraries of mouse and human clones, which we then spot on glass slides. We also provide printing services for Einstein researchers who wish to print PCR products from other genomes. Weíve spotted custom RNA chips, chips with all TB genes, mouse neural chips, chromosome-specific chips, and others. Users have a choice of using either our in-house chips or any of the Affymetrix GeneChip arrays. The majority of users prefer the in-house chips because of lower costs and the ability to customize the gene set.
Q Have you developed any special protocols for optimization of array performance?
AShufen Chen, a talented technician in [Einstein molecular genetics researcher] Geoffrey Childsí lab, developed and optimized a more economical high-throughput method for preparation of high-quality PCR products. We also modified the hybridization protocol to make the process more consistent across many users and allow use of either polylysine- or amino silane-coated slides.
Q What kind of arrayer do you use to make your own arrays?
AThe printer that I designed and built allows us to print more than 30,000 spots per slide using 48-quill pins onto 230 slides at a time. It uses PCR products very efficiently so that our price-per-chip is quite reasonable.
Q What methods do you use to analyze microarray data?
AIn the beginning, we exclusively used Mike Isenís ScanAlyze and Cluster with Treeview. As the commercial products got better, some labs began using them. Some developed their own Excel macros or Access databases for automating data analysis. A year ago, we decided to purchase the Axon scanner, which comes bundled with the GenePix gridding software. Now the higher-volume users are using GenePix and some are using GeneSpring for further analysis.
Q What is the biggest challenge you face in working with microarrays?
AOne, getting enough pen tips that print reliably and two, users who blame the chip when things donít work.
QHow do you tackle these challenges?
AThe hardware problems are handled by standard engineering procedures. The user issues require an ongoing training program for new users, and interaction with the facility staff when problems appear.
Q If you could make out a wish list for microarray technology advances or improvements over the next couple of years, what do you most want or need?
AFirst, fully annotated gene libraries. Second, standards for data interchange among various facilities and platforms. Third, better bioinformatics tool sets. And finally, better protocols that allow use of very small quantities of RNA and yield reliable results.