Gerald J Wyckoff Training modes Are we training genomicists to deal with the right problems Are the training modalities were using proper What questions should we be looking at Problems in Drug Discovery ID: 932691
Download Presentation The PPT/PDF document "The next generation of genomicists" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
The next generation of genomicists
Gerald J. Wyckoff
Slide2Training modesAre we training
genomicists to deal with the right problems?Are the training modalities we’re using proper?What questions should we be looking at?
Slide3Problems in Drug Discovery
FDA protocols.Success in animals before phased clinical drug trials**Most successful drugs target very conserved evolutionarily, probably in part for this reason
Personalized medicine
(Disclosure: I’m the founder of
Zorilla
Research, commercializing software from my lab for Drug Discovery and
OneHealth markets)
Slide4Drug Discovery
While costs for many technologies are dropping quickly…The latest report from the Tufts CSDD says costs for bringing a successful drug to market are $2.5B USD
$1B USD is prior to clinical trials
Nature, 9 March 2014
Slide5The next generation of genomicists
Embrace the need for new modes of thinkingPut genomics and evolutionary genomics at the center of drug development and discovery
Develop new models
Slide6Current Approach is Wrong
Our In silico work needs to go beyond docking to look at picking targets in an intelligent manner
Not all targets are going to allow successful development
An overwhelming number of Human alleles were suboptimal
1
Proteins that are the targets off successful drug development are very conservative- even though disease genes are often very rapidly evolving
2
Recently, data showed that common variants- even in supposedly “normal” individuals- may play a roll in some
multigenic
diseases
3
Not all disease genes are potential targets
Taken together, this means that there are many disease causal or contributing alleles (gene variants) in the population that fall below the threshold of interest for drug discovery efforts
But a precision medicine approach, which is where the industry is heading, will require drugs against a broad number of pre-existing conditions; not something the current drug pipeline is geared to do
1
Fay JC, Wyckoff GJ, Wu CI. Positive and negative selection on the human genome.
Genetics
. 2001;158(3):1227-1234.
2
Qing Zhang, Ada
Solidar
, Nicholas J.
Murgolo
,
Wynand
Alkema, Wei Ding, Peter M.
Groenen
, Jonathan R. Greene, Eric L. Gustafson,
Jan
Klomp
, Ellie D. Norris, Ping
Qiu
and Gerald J. Wyckoff, 2012. Selective Constraint: A Hallmark of Genes Successfully Targeted for Pharmaceutical Development.
American Journal of Drug Discovery and Development, 2: 184-193.
3
Ubadah
Sabbagh
,
Saman
Mullegama
, and Gerald J. Wyckoff, “Identification and Evolutionary Analysis of Potential Candidate Genes in a Human Eating Disorder,”
BioMed
Research International, vol. 2016, Article ID 7281732, 11 pages, 2016. doi:10.1155/2016/7281732
Slide7Not Just Docking
The “disease first” approach for Drug Discovery to date WILL NOT suffice in the very near futureWe need to be intelligently choosing eventual targets, not just potential targets
Slide8Where do we go from here?
Slide9Many deleterious SNPS
Estimates show that each individual carried far more deleterious SNPs (Single Nucleotide Polymorphisms) than are identified as “disease causal”
These are contributing to “genetic load”; another way of looking at it is each of these genes may manifest as a disease, disorder, or proclivity towards disease under certain environmental conditions
These aren’t what we typically think of as “disease causal”, because in many individuals they don’t actually cause disease- because the person wasn’t exposed to a particular environmental factor
Slide10Measuring evolution
We use the synonymous (non-amino acid changing) nucleotide changes as a ruler for how much change would take place if there was no selection.We would expect (under neutrality) that there would be as many amino acid changes.
Correct for number of sites.
Called
Ka
/Ks.
Slide11Conservative Bias
Qing Zhang, Ada
Solidar
, Nicholas J.
Murgolo
,
Wynand Alkema, Wei Ding, Peter M. Groenen
, Jonathan R. Greene, Eric L. Gustafson, Jan
Klomp
, Ellie D. Norris, Ping
Qiu
and Gerald J. Wyckoff, 2012. Selective Constraint: A Hallmark of Genes Successfully Targeted for Pharmaceutical Development. American Journal of Drug Discovery and Development, 2: 184-193.
Slide12Can we actualize this?We wanted to know if we could put this into practice using data from public sources
We were interested in the link between sleep and metabolic disordersNotably, the link between metabolism and neurological function in obesity
Slide13Example with GEO
Ubadah
Sabbagh
,
Saman
Mullegama
, and Gerald J. Wyckoff, “Identification and Evolutionary Analysis of Potential Candidate Genes in a Human Eating Disorder,”
BioMed
Research International, vol. 2016, Article ID 7281732, 11 pages, 2016. doi:10.1155/2016/7281732
Slide14Results
1,052 genes significantText analysis: about 100 interestingSeveral overlap across all sets:
Slide15From Text Analysis:
Slide16Evolutionary analysis: Problem
Slide17What’s Normal?
Some SNPs identified as “normal” even in reference populations likely aren’t
We see that in a screen looking at potential genes involved with eating disorders.
Some genes have a clear excess of SNPs predicted to be deleterious by programs such as SIFT or
Polyphen
Slide18Common Disorder, Common Variant
“The common disease-common variant (often abbreviated CD-CV) hypothesis predicts that common disease-causing alleles, or variants, will be found in all human populations which manifest a given disease. Common variants (not necessarily disease-causing) are known to exist in coding and regulatory sequences of genes. According to the CD-CV hypothesis, some of those variants lead to susceptibility to complex polygenic diseases. Each variant at each gene influencing a complex disease will have a small additive or multiplicative effect on the disease phenotype. These diseases, or traits, are evolutionarily neutral in part because so many genes influence the traits.”
Slide19Back to the future!
Wyckoff GJ, Wang W, Wu CI. Rapid evolution of male reproductive genes in the
descent of man. Nature. 2000 Jan 20;403(6767):304-9. PubMed PMID: 10659848.
Slide20Where do we go from here?
Better integration between Human and animal data at all levels of drug discovery.Pathway analysisPK/PD analysis
Don’t lose data from failed drug studies.
Evolutionary analysis is key -
OneHealth
Need new tools for aggregating target information.
Help prevent late stage failures in drug development.
Big Data Approach
Slide21Training the future
Must utilize new modes to ensure integrated learning across all genomics-related disciplinesContinuing education opportunities for professionals in fields related to pharmaceuticals, medicine“Borrow” OneHealth
concepts- work with Veterinarians.
Slide22For Further Information, contact:
wyckoffg@umkc.edu
Acknowledgments
The Wyckoff Lab
Lee Likins, Scott Foy, Ming Yang, Andrew
Skaff
,
Ubadah
Sabbah
,
Saman
Mullegama
, Adam
Younkin
Ada
Solidar
(B-tech Consulting)
The
Miziorko
Lab (UMKC)
Jeff Murphy (Nickel City Software)
Brian
Geisbrecht
(KSU)
And his lab
John Walker (SLU)
Jim
Riviere
(KSU)
Tom
Menees
(UMKC)
Zorilla
Research
John Crowe, Lee Whittaker
NIH 1 R41 GM 088922-01A1
NIH 2 R44 GM097902-02A1 (
Dockhorn
, PI)
NIH 1 R21 AI113552-01 (
Geisbrecht
, PI)
VaSSA
Informatics, LLC for major funding
Digital Sandbox KC
Missouri Technology Corporation
UMKC Fast-Track
UMKC SBS, UMRB, UMKC FRG, KCALSI for additional funding