- Ph.D. Biochemistry , Washington University 2007.
Thesis : Mimicking protein reverse turns with cyclic tetrapeptides - computational prediction and synthetic validation.
- BS Chemistry, BS Biology (biochemistry concentration) , Duke University 2000.
My interests started with nanotechnology and have evolved to include biotechnology, computational biology, and public policy. While working over four years at the National Institute of Standards and Technology (NIST) on improving Scanning Tunneling Micscroscopy (STM), I started to see proteins as natural nanomachines and attained my BS in chemistry and biology with a minor in biochemistry from Duke University. Following college I worked in a startup biotech helping to sequence the first human genome for the NIH, after which I went on to attain my Ph.D. in biochemistry from Washington University School of Medicine where I researched protein structure and developed rigid compounds to act as therapeutics. Proceeding my thesis work I delved into various field in industry starting as a systems biologist at Pfizer elucidating molecular pathways, drug targets, and describing clinical trial population results due to genomic variation. I then managed an organic and microbial lab at Eurofins Scientific, and most recently led the global development of a new LIMS at Pioneer Hi-Bred architecting databases and analytical tools to help with genotyping. My research aims to improve the human condition through drug design and compound repurposing while studying protein misfolding, epigenetics, HIV, cancer, and causal links found via mining datasets.
With training in computational structure prediction, molecular and cellular biology, organic chemistry, data mining, and public policy my research combines the following areas of interests & methods of investigation in an effort to maximize my contribution to scientific enlightenment.
|Areas of Interest ||Methods |
|Alzheimers ||Conformational Search , Dynamics, Docking |
|HIV ||Library synthesis |
|Epigenetics ||Data mining |
|Viruses ||Membrane modeling |
|Water , Protein , Structure ||Binding assays |
|Public Policy ||Genetic modification |
There will be ample projects available for student engagement with little need for prior training. For example, computational structural biology can be picked up in as little as a week if working on a very focused topic. Many of the tools one would learn engaging in my research prove useful in a plethora of functions after becoming a doctor. For example having been involved in data mining will help catapult a doctor into the select echelon of those that can interact with the deluge of patient data that physicians will have to deal with in the 21st Century. Imagine the relative effectiveness between two physicians, when one will incorporate a patient's genomic, epigenetic, and environmental health data while the other will not. While much of this analysis seems daunting at first, there is an early learning curve barrier, which once overcome allows for easy, routine use of such data.
Alzheimers Amyloid Binding Therapeutic
Research interests include computational design of amyloid binding entities. Targeting monomer, dimers, and up to octomers of amyloid aggregated proteins.
Just over 100 years ago Alois Alzheimer used staining techniques and described the pathology and symptoms of dementia that would come to be called Alzheimer’s Disease (AD). An increasing percentage of the world population will get AD in their lifetime with 7% of those over 65 and 40% of people over 80 years of age in industrialized countries being affected (Glass 2010, Cell). Despite decades of AD research an effective therapy let alone a vaccine has not been developed. However, the wealth of molecular information now known about the AD progression, and recent leads, raise hopes for the next decade of therapeutic development.
Multiple root causes of Alzheimer’s dementia have been proposed, which have varying levels of data behind them. The amyloid theory, for decades the most widely accepted paradigm, initially posited that the amyloid plaques kill brain cells and are the causative agent leading to Alzheimer’s dementia. In 1995 genetic links leading to Early onset Familial Alzheimer's disease (EOFAD) were found (Glass 2010, Cell). These mutations were in the presenilin gene which is a component of γ-secretase which cleaves the amyloid precursor protein (APP) releasing the amyloid peptide to the cytosol leading to the formation of amyloid plaques (Figure 1). In mouse models mutation of the presenilin gene causes Alzheimer’s like symptoms and could be alleviated through inactivation of presenilin (Clark 1995 , Nature). Multiple methods have proven efficacious in reducing amyloid plaques and some have been correlated with better cognitive outcome while others have not.
While both amyloid plaques and tau tangles are macroscopically visible phenomena that correlate with AD progression, recent research suggests they may not be the causative agent of dementia (Clark 1995 , Nature). A new model to fit available data has been put forward in which it is amyloid-β (Aβ) pre-plaque monomer or small oligomer that cause neuron death, not the larger plaque (Saura 1995 , J Neuroscience). In this model amyloid plaques could even inhibit neuron death by sequestering the deleterious cytolsolic amyloid monomers. In this light, previous researchers conclusions about amyloid plaque may be reinterpreted as correlative instead of causative. Compounds may have been measured to decreases amyloid plaques and result in delaying cognitive decline, however this could be due to decreased soluble amyloid which in turn decreased the plaques. While there is not consensus on the causative agent leading to AD neuron death and dementia, inhibiting Aβ in the cytosol has a significant track record already as a therapeutic target with mixed results.
Amyloid dimer with ligands bound
Amyloid cleavage pathway
HIV drugs made by bread
My research will elucidate if the following three scientific discoveries can be combined to create HIV therapeutics which can be manufactured by yeast or bacteria and delivered by bread or yogurt respectively.
- HIV is now known to bind the coreceptor CXCR4 often during the transition to the more aggressive stage of AIDS.
- Split inteins are two proteins which splice the peptides attached to their ends together when they come in contact. It has been shown that putting a peptide as small as 4 amino acids in between these split inteins can cause cyclization of the 4 residues.
- Cyclic tetrapeptides have been shown to
- mimic over half the reverse turns in proteins
- be bioavailable as therapeutics
Using cyclic tetrapeptides to mimic reverse turns at the interface to CXCR4 and HIV could result in a therapeutic that can be delivered in food made with genetically modified organisms. This would bring the cost of the therapeutic to nothing after development and help in delivery to the third world.
Engineered nanocomputers in a cell
It should now be possible to engineering proteins in a cell with unique amino acid codes to be recorded by the cell as they interact with other proteins. If we genetically modify the below constructs :
It should be possible when they interact to result in the following protein:
You could imagine after interacting with several proteins over time you would end up with:
At the end of an experiment reading the protein sequence (by Edmond degradation for example) would provide a list of the interactions this protein had interacted with. Instead of getting snapshots during an experiment this would be a complete history of interactions during the experiment. Protein99 could for example be a non-interacting protein used as a control. Its incorporation would serve as a time domain, being incorporated at random intervals based on concentration. In the above example you could see that protein1 interacted more quickly with proteins 2, 86, 23, and 29 than it did with 73.
Epigenetic histone binding therapies
There has been significant press over the plummeting costs of genotyping, and rightfully so. In the next 15 years the cost of sequencing your entire genome should drop to the price of a cup of cappuccino, be free, or even be a product you get paid to allow someone to perform and access. However, less light has been shed on the ramifications of the epigenetic information that will be coming. Epigenetic data regards the turning on and off of genes in your DNA. While you could have your DNA sequenced once and have that data be correct for the rest of your life (unless you develop cancer), your epigenome could be “resequenced” at every doctors visit and change. In fact, reading someone’s epigenetics could be the best way to see if treatments are working. Prescribing exercise or a nutrition system could then be monitored by seeing if certain deleterious genes are turned off epigenetically. Admittedly the knowledge and low cost of epigenetic information is far behind that of genomic data, but it will no doubt follow quickly. The tools used in computational drug design, are now being turned to create therapeutics to effect someone’s epigenetics. Namely the methylation of histones can be targeted by rationally created therapeutics. My research interests in this area start at targeting histone deacetylase , largely because they are the best described targets currently.
Water structure epiphanies effecting batteries , clouds, and potable water
There is a layer of water molecules that surround hydrophilic (water loving) surfaces in an ordered manner. This has been known and accepted for decades, and the design of drugs has taken this into account since the exterior of proteins can have this layer of ordered water molecules. Recently however, evidence has been presented (Pollack, U. Washington) which suggests that instead of this layer consisting of a few water molecules in depth it could in fact order thousands of water molecules. Such a situation can be seen at a macroscopic level, and if these results hold up have profound repercussions. In essence this pure ice-like layer of water that would form (almost a millimeter thick) could provide purified drinking water for example. In addition this layer has a negative charge and absorbs in the infrared spectrum and could therefore be used as an energy source rivaling any current renewable energy in economic viability. This Exclusion Zone (EZ) structure of water needs to be investigated, validated, and modeled.
Cures to viruses – restriction enzymes
Infectious diseases can sadly still be largely split into bacterial (treatable with antibiotics) and viral (untreatable except in rare cases). Once a virus integrates into a person’s DNA they are currently left with managing the side effects for the rest of their life. Technology exists to now custom design restriction enzymes (which cut DNA) to specific genomic sites. While there are many details to iron out (e.g. delivery mechanism, low excise error rate, effective ligation), I have no doubt we will be designing therapeutics to excise viral DNA and would like to be part of this research.
We are at a unique point in time, where more data is being gathered and stored digitally than we can currently analyze. While most research aims at forming a thesis and then creating and analyzing all necessary data, there is a new breed of scientist (e.g. Atul Butte at Stanford) that uses publically available datasets to data mine interesting information. This is the most high-tech form of recycling and can yield amazing results. With a SQL database background and computational scripting skills combining divergent datasets seem particularly interesting. For example imagine the correlations between data from : genomic chips, sleep pattern recording device, GPS/heart rate fitness apps, and nutrition facts stored from a membership card.
See some interesting data below to put the state of the US healthcare in reference to other developed countries … there is room for improvement, and I look forward to working with my colleagues and students to make up loss ground.