IOM report tackles -omics validation for clinical studies after Duke scandal

March 23, 2012 § 3 Comments

The fallout from the infamous Duke University clinical trial scandal with Anil Potti and Joseph Nivens’ work on genomic signatures hasn’t been pretty, and the Institute of Medicine wants to make sure it doesn’t happen again. Today, the institute released its report called Evolution of Translational Omics: Lessons Learned and the Path Forward.

The report discusses how to make sure discoveries made in labs using -omics technologies can be properly validated and confirmed before the findings are turned into clinical therapies or diagnostics tests and used on patients. The report points out that the road to turn -omics-based discoveries into clinically useful tests has been more challenging than anticipated or appreciated.

The press release notes:

Omics-based tests involve large data sets and complex algorithms, and investigators do not routinely make their data and computational procedures accessible to others who could independently verify them.  The regulatory steps that investigators and research institutions should follow may be ignored or misunderstood.  As a result, flaws and missteps can go unchecked.

Why and how is -omics data validation a different beast from validation of other scientific findings? In a nutshell, it’s all about the sheer volume of information produced by -omics technologies, such as microarrays and mass spectrometry. When the number of data points start hitting the millions, it becomes a new paradigm in how these data can be checked for soundness.

But how to implement ways to double-check that investigators, starting at the fundamental research level all the way to clinical trials, aren’t chasing ghosts in data is easier said than done. It needs validation to be incorporated into the discovery process, unlike now when validation is done after a discovery is made. Making that change is challenging.  As the report points out, the change will require cooperation between entities like the research community, federal funding agencies and regulatory agencies like the U.S. Food and Drug Administration.  The report makes recommendations for best practices to ensure data are valid.

Be sure to check out the April 2012 issue of ASBMB Today, which will be available online by Monday, April 2. In that issue, I have an in-depth feature article that delves into the challenges of validating -omics data. One of the people I interviewed for the story is Gil Omenn at the University of Michigan, who chaired the IOM committee.

Stay tuned…

§ 3 Responses to IOM report tackles -omics validation for clinical studies after Duke scandal

  • Cindy James says:

    I have been a biochemist for 23 years and work with a top notch team of Mass Spectrometry and Proteomics people at a major university who are extremely diligent about making sure that the data is perfect. An often heard phrase is “We can’t make this sh_t up!”

    Many times results are not what the investigator wants to see. We have been asked to “turn the sensitivity down” on the instrument or “erase” a line that represents a contaminant. I have even been asked to photoshop an image by a grad student because the band on the SDS-PAGE gel appeared on both his control AND his sample!

    Many young researchers lack the biochemistry knowledge to make prudent decisions about their projects and often rely on purchased “kits” to conduct their work, not understanding the chemistry involved. We have people who make errors in their work and their supervisors do not catch them….I do!

    Supervisors, principle investigators and clinicians need to make an appearance in the laboratories every now and then and demand to view and then to completely understand the raw data. Many do not.

    Research is one area of work which demands us to be as suspicious and critical of each other as we possibly can. This includes hiring people who have profiles “online”. Even my 5 year old grandson can make a website…….

  • Raj Mukhopadhyay says:

    Jocelyn Kaiser has written an excellent piece for ScienceInsider to explain what the report is all about and how it came about. Check it out at:

  • Raj Mukhopadhyay says:

    And here it is: My feature story on validating -omics data. Let me know what you think!

Leave a Reply

Your email address will not be published. Required fields are marked *

What’s this?

You are currently reading IOM report tackles -omics validation for clinical studies after Duke scandal at Wild Types.