{"response":{"award":[{"abstractText":"The broader impact/commercial potential of this Small Business Innovation Research Phase II project will combine computational protein engineering (CPE) software, tools and methods into a \"platform technology\" that enables new products and technologies in a wide range of scientific areas, including industrial enzymes, pharmaceuticals, therapeutics, medical diagnostics and bioenergy. This project will validate the CPE platform in an application that has both commercial and environmental value: engineered enzymes for utilizing methane as a feedstock. The engineered methane activating enzymes developed will open up numerous new biological routes to chemicals and fuels from natural gas, capitalizing on abundant domestic shale gas reserves and improving U.S. energy independence. This CPE enzyme technology has the potential to significantly decrease the capital cost of small-scale Gas-to-liquid (GTL) plant, enabling the capture and monetization of otherwise economically-stranded natural gas. In North Dakota, more than a quarter of the total gas produced is flared or vented, wasting a valuable natural resource and needlessly emitting greenhouse gases. By facilitating the conversion of stranded or flared methane to fuels and high-value chemicals, this research can help reduce carbon footprint and spur domestic manufacturing, investment, and job creation. \r\n\r\nThe objectives of this Phase II project are to develop and license a CPE Platform technology that will enable experimental biological scientists to harness the power of computational protein design and engineering; to validate the CPE Platform technology by engineering novel methane activating enzymes that have increased soluble expression and methane-oxidizing activity; and in collaboration with potential licensees of this technology, work to incorporate the engineered enzyme into their industrial host organisms. These organisms will be enabled to use methane as a feedstock for the bio-production of a wide range of end products, including liquid transportation fuels, commodity chemicals and high value fine chemicals. In Phase I, the Northwestern-Caltech-Protabit team succeeded in engineering the particulate MMO (pMMO) catalytic subunit (spmoB) to allow it to be solubly expressed in E. coli. This was the first demonstration of an active methane-oxidizing enzyme that can be solubly expressed and purified in significant quantities in a genetically-tractable recombinant host. Phase II will continue this work, with the objective of making the spmoB enzyme more effective by increasing its activity and solving certain other challenges for inserting it into an industrial host.","activeAwd":"false","agency":"NSF","awardAgencyCode":"4900","awardee":"PROTABIT, LLC","awardeeAddress":"111 WAVERLY DR","awardeeCity":"PASADENA","awardeeCountryCode":"US","awardeeDistrict":"28","awardeeDistrictCode":"CA28","awardeeName":"Protabit LLC","awardeePhone":"6268447348","awardeeStateCode":"CA","awardeeZipCode":"911052513","cfdaNumber":"47.084","coPDPI":["Stephen L Mayo steve@mayo.caltech.edu"],"date":"09/14/2015","dirAbbr":"TIP","divAbbr":"TI","estimatedTotalAmt":"750000","expDate":"08/31/2018","fundAgencyCode":"4900","fundProgramName":"STTR Phase II","fundsObligated":["FY 2015 = $750,000.00"],"fundsObligatedAmt":"750000","histAwd":"false","id":"1534743","initAmendmentDate":"09/14/2015","latestAmendmentDate":"01/14/2016","managingPec":"159100","orgCodeDir":"15000000","orgCodeDiv":"15030000","orgLongName":"Directorate for Technology, Innovation, and Partnerships","orgLongName2":"Translational Impacts","orgUrl":"https://beta.nsf.gov/tip/ti","parentUeiNumber":"","pdPIName":"Barry D Olafson","perfAddress":"1200E California Blvd","perfCity":"Pasadena","perfCountryCode":"US","perfDistrict":"28","perfDistrictCode":"CA28","perfLocation":"California Institute of Technology","perfStateCode":"CA","perfZipCode":"911250001","pi":["Barry D Olafson barry.olafson@protabit.com"],"piEmail":"barry.olafson@protabit.com","piFirstName":"Barry","piId":"269883529","piLastName":"Olafson","piMiddeInitial":"D","poEmail":"","poName":"Anna Brady-Estevez","poPhone":"","primaryProgram":["01001516DB NSF RESEARCH & RELATED ACTIVIT"],"progEleCode":"159100","program":"STTR PHASE II, Chemical Technology, Biotechnology","progRefCode":"1591, 8030, 8038","projectOutComesReport":"
\n

The field of protein engineering and protein drug discovery is undergoing a transformation enabled by the use of AI and machine learning approaches, most recently deep neutral networks, that correlate changes in protein sequence to protein function. The large amounts of protein sequence mutation data that is needed to train machine learning models is being generated at ever faster rates through deep sequencing experiments, robotics and lab automation. Our ProtaBank database facilitates the collection, validation and organization of protein sequence mutation data over a wide range of properties.

\n

This STTR project has been instrumental in the development of two computational protein deign (CPD) software platforms. The first platform, the ProtaBank AI Platform shown in Figure 1, integrates the ProtaBank database and a number of available AI frameworks that provide a wealth of machine learning tools to use in training models to optimize beneficial properties of proteins upon sequence mutation. These learned models could then be used to engineer new protein variants that will become tomorrow?s therapeutic proteins or industrial enzymes. These AI tools become increasingly effective as more protein domain knowledge is imparted to the training process. For example in the current STTR project, we developed multiple sequence alignment (MSA) algorithms, and Partial Order Optimum Likelihood (POOL) calculations to create new encoding schemes with enriched calculated protein properties to enhance the data that is presented to the machine learning module in the ProtaBank AI Platform. As new approaches like this become available through publication in the scientific literature, they can be incorporated into the ProtaBank AI Platform, allowing users of the ProtaBank AI Platform greater access to the most current tools they will need to perform their protein engineering problems.

\n

The second computational product that was enhanced by the work of this STTR project is our TRIAD Computational Protein Engineering (CPE) Platform. This platform incorporates many of the more traditional CPD algorithms into a physics-based design of beneficial protein mutations. This platform works with other experimental approaches like site saturation mutagenesis, chimeragenesis and directed evolution experiments to suggest new pathways for engineering proteins. It complements the new emerging AI approaches in that it calculates the new combinatorial protein libraries that will serve as the basis for generating problem-specific protein sequence mutation data for use in the ProtaBank AI Platform. TriadWeb, the web-based version of the TRIAD CPE Platform, brings the power of CPD to the desktop and provides simple access and ease-of-use to protein engineers who want to use design tools in the course of their engineering projects. A sample input screen for TriadWeb is show in Figure 2.


\n

\n\t\t\t\t \tLast Modified: 12/26/2018
\n\t\t\t\t\tModified by: Barry D Olafson

\n
\n
\n
\n
\n
\n\t\t\t\t\t\t\t\t\tImages (1 of )\t\t\n\t\t\t\t\t\t\t\t
\n
\n
\n
\n
\n
\n
\n
\n
    \n
  • \n\"The\n
    \n
    Figure 1. The ProtaBank AI Plaftorm
    \n
    Protabit LLC
    \n
    Barry D Olafson
    \n
    The ProtaBank AI Platform
    \n
    \n
    • \n
  • \n\"TriadWeb\n
    \n
    Figure 2
    \n
    Protabit LLC
    \n
    Creative Commons
    \n
    Barry D Olafson
    \n
    TriadWeb Sequence Design
    \n
    \n
    • \n
\n
\n
\n
\n
","publicAccessMandate":"0","startDate":"09/15/2015","title":"STTR Phase II: Development of a computational protein engineering platform and its application to methane activating enzymes","transType":"Standard Grant","ueiNumber":"ZMG1FJD9SQ64"}],"metadata":{"offset":0,"rpp":25,"totalCount":1}}}