SAS, JMP and the JMP Genomics team hosted the seventh face-to-face meeting of the MicroArray Quality Control (MAQC) consortium May 24-25 on the SAS campus. Over the past few years, the MAQC has attracted significant attention from thought leaders in biotech and pharma, chip manufacturers, and analytics providers, including developers and testers from SAS and in particular members of the JMP Genomics team led by Russ Wolfinger.
SAS executive vice-president and JMP developer John Sall kindly provided opening remarks for the MAQC meeting on Thursday morning, commenting on the long history of SAS in the life sciences. He also shared some interesting tidbits about the history of SAS and, in particular, of Building V, where the meeting was held, which formerly housed both video game development and production of exercise infomercials! Maybe I’m just still a newcomer to SAS, having been here only a year, but those chapters in Building V history were news to me…
Attendees of this MAQC meeting have been asked to keep content of scientific presentations confidential since they will be the basis for future publications. However, the goals of the MAQC are certainly not secret in any way. They’re clearly stated on the FDA home page for the MAQC: “The purpose of the MAQC project is to provide quality control tools to the microarray community in order to avoid procedural failures and to develop guidelines for microarray data analysis by providing the public with large reference datasets along with readily accessible reference RNA samples.”
The JMP Genomics team and others from SAS participated in several papers from Phase I of MAQC published in Nature Biotechnology last September, and are also involved in analysis efforts for Phase II data sets. One general goal of Phase II is to offer guidance on methods for building robust predictive models with genomics data. Coupled with demand from users for expanded predictive modeling in JMP Genomics, this MAQC goal helped drive development of greatly expanded predictive modeling capabilities for JMP Genomics 3.0. The development team added a Radial Basis Machine process and the Cross-Validation Model Comparison platform for easily comparing the performance of large groups of predictive models (both models of the same type and across different types). It’s become quite clear over the past few months that these capabilities set JMP Genomics apart among our competition in the genomics analysis space.
From my perspective, MAQC participants have great scientific interests in the statistical techniques and methods which can be used to analyze genomics data, but also a genuine desire to see good quality genomics data be used to benefit human health and support the drug development process. With the introduction of voluntary submission of genomics data (VGDS) into the drug approval process, the FDA clearly desires more rigorous exploration and standardization of analytic processes for the analysis of genomics data. The FDA coordinators of MAQC have stepped in to facilitate research in that area, and scientists and software developers alike have a lot to learn from their findings and publications thus far.
One particularly exciting public development at the MAQC meeting here at SAS was the announcement of an MAQC Whole Genome Association (WGA) study group. With JMP Genomics 3.0, we’re supporting analysis of some very large WGA datasets. We’ve recently benchmarked a number of our genetics processes with 500k SNP genotypes for 4000 samples, and some with up to a million SNPs for each of 4000 samples. I have no doubt that the SAS and JMP Genomics developers will be heavily involved in the future direction of the WGA study group, and will incorporate what they learn into future improvements to the amazing genetics analysis capabilities in JMP Genomics.
My personal thanks to John Sall, JMP Marketing, and all others from SAS who helped plan and support this important two-day event!