Matthew A. Kayala

PhD Candidate Email : mkayala [at] ics [dot] uci [dot] edu Phone : (401) 339-7392 Office: 243 ICS2 (map) Address: Bren School of Information and Computer Sciences University of California, Irvine CA 92697 Open source code: Github   I am a PhD candidate advised by Pierre Baldi at the University of California, Irvine in the Bren School of Information and Computer Sciences. I am associated with the Institute for Genomics and Bioinformatics. Broadly, my research involves using statistical machine learning to explore and understand biological and chemical phenomena. I hold MS and BS degrees in Computer Science from the University of Rhode Island. My full CV can be found here.

Current Research

• Prediction of chemical reactivity:
  State of the art systems to predict chemical reactivity are either not high-throughput, i.e., saddle-point methods with QM approximations, or are brittle and manually encoded rule-based systems. I work on a new machine learning approach to reaction prediction. This approach uses statistical ranking models learned from databases of known synthetically productive reactions. A web server predictor for a subset of chemistry can be found here.
• Differential analysis of high-throughput data:
  A common task in the analysis of data generated by high-throughput biological techniques (such as DNA microarrays or Mass Spectrometry) is determining which measurements are statistically different between two or more conditions. This task is often undertaken in low replicate regimes. I have helped to extend a Bayes-regularized approach to this problem. A webserver implementing these methods can be found here. I have used these methods (amongst others) in collaborations with several immunology groups to analyze protein microarray data.
• Metabolic site prediction:
  Some of the methods I have developed for chemical reactivity prediction are directly applicable to the specific problem of regioselectively predicting sites of cytochrome P450 (CYP) mediated metabolism. Methods to predict the localized sites (and mechanisms) of this metabolism are necessary components of a medicinal chemistry toolbox for lead optimization.

Peer-Reviewed Journal Publications

• M.A. Kayala and P. Baldi.
  Cyber-T webserver: Differential analysis of high-throughput data
  Nucleic Acids Research Web Server Issue (In Press), (2012).
  
  
• S-J. Lee, L. Liang, S. Juarez, M.R. Nanton, E.N. Gondwe, C.L. Msefula, M.A. Kayala, F. Necchi, J.N. Heath, P. Hart, R.M. Tsolis, R.S.Heyderman, C.A. Maclennan, P.L. Felgner, D.H. Davies, and S.J. McSorley
  Identification of a common immune signature in murine and human systemic Salmonellosis.
  PNAS, 109, 13, 4998-5003, (2012).
  (Pubmed) (pdf)
  
  
• M.A. Kayala, C.A. Azencott, J. Chen, and P. Baldi.
  Learning to predict chemical reactions.
  Journal of chemical information and modeling, 51, 9, 2209-2222, (2011).
  (Pubmed) (pdf)
  
  
• A.E. Barry, A. Trieu, F.J.I. Fowkes, J. Pablo, M. Kalantari-Dehaghi, A. Jasinskas, X. Tan, M.A. Kayala, L. Tavul, P.M. Siba, K.P. Day, P. Baldi, P.L. Felgner, and D.L. Doolan.
  The stability and complexity of antibody responses to the major surface antigen of Plasmodium falciparum are associated with age in a malaria endemic area.
  Molecular and cellular proteomics, EPub. M111.008326, (2011).
  (Pubmed) (pdf)
  
  
• A. Trieu, M.A. Kayala, C. Burk, D.M. Molina, D.A. Freilich, T.L. Richie, P. Baldi, P.L. Felgner, D.L. Doolan.
  Sterile protective immmunity to malaria is associated with a panel of novel P. falciparum antigens.
  Molecular and cellular proteomics, 10, 9, M111.007948, (2011).
  (Pubmed) (pdf)
  
  
• C.N. Magnan, M. Zeller, M.A. Kayala, A. Vigil, A. Randall, P.L. Felgner, and P. Baldi.
  High-throughput prediction of protein antigenicity using protein microarray data.
  Bioinformatics (Oxford, England), 26, 23, 2936-2943, (2010).
  (Pubmed) (pdf)
  
  
• M.I. Cruz-Fisher, C. Cheng, G. Sun, S. Pal, A. Teng, D. Molina, M.A. Kayala, A. Vigil, P. Baldi, P.L. Felgner, X. Liang, and L.M. de la Maza.
  Identification of immunodominant antigens by probing a whole Chlamydia ORFome microarray using sera from immunized mice.
  Infection and immunity, 79, 1, 246-257, (2010).
  (Pubmed) (pdf)
  
  
• P.D. Crompton, M.A. Kayala, B. Traore, K. Kayentao, A. Ongoiba, G.E. Weiss, D.M. Molina, C.R. Burk, M. Waisberg, A. Jasinskas, X. Tan, S. Doumbo, D. Doumtabe, Y. Kone, D.L. Narum, X. Liang, O.K. Doumbo, L.H. Miller, D.L. Doolan, P. Baldi, P.L. Felgner, and S.K. Pierce.
  A prospective analysis of the Ab response to Plasmodium falciparum before and after a malaria season by protein microarray.
  PNAS, 107, 15, 6958-6963, (2010).
  (Pubmed) (pdf)
  
  
• D.M. Molina, S. Pal, M.A. Kayala, A. Teng, P.J. Kim, P. Baldi, P.L. Felgner, X. Liang, and L.M. de la Maza.
  Identification of immunodominant antigens of Chlamydia trachomatis using proteome microarrays.
  Vaccine, 28, 17, 3014-3024, (2010).
  (Pubmed) (pdf)
  
  
• A.B. Mochon, J. Ye, M.A. Kayala, J.R. Wingard, C.J. Clancy, M.H. Nguyen, P.L. Felgner, P. Baldi, and H. Liu.
  Serological profiling of a Candida albicans protein microarray reveals permanent host- pathogen interplay and stage-specific responses during candidemia.
  PLoS pathogens, 6, 3, e1000827, (2010).
  (Pubmed) (pdf)
  
  
• L. Liang, D. Leng, C. Burk, R. Nakajima-Sasaki, M.A. Kayala, V.L. Atluri, J. Pablo, B. Unal, T.A. Ficht, E. Gotuzzo, M. Saito, W.J.W Morrow, X. Liang, P. Baldi, R.H. Gilman, J.M. Vinetz, R.M. Tsolis, and P.L. Felgner.
  Large scale immune profiling of infected humans and goats reveals differential recognition of Brucella melitensis antigens.
  PLoS neglected tropical diseases, 4, 5, e673, (2010).
  (Pubmed) (pdf)
  
  
• P.L. Felgner, M.A. Kayala, A. Vigil, C. Burk, R. Nakajima-Sasaki, J. Pablo, D.M. Molina, S. Hirst, J.S. Chew, D. Wang, G. Tan, M. Duffield, R. Yang, J. Neel, N. Chantratita, G. Bancroft, G. Lertmemongkolchai, D.H. Davies, P. Baldi, S. Peacock, and R.W. Titball.
  A Burkholderia pseudomallei protein microarray reveals serodiagnostic and cross-reactive antigens.
  PNAS, 106, 32, 13499-13504, (2009).
  (Pubmed) (pdf)
  
  
• A.G. Barbour, A. Jasinskas, M.A. Kayala, D.H. Davies, A.C. Steere, P. Baldi, and P.L. Felgner.
  A genome-wide proteome array reveals a limited set of immunogens in natural infections of humans and white-footed mice with Borrelia burgdorferi.
  Infection and immunity, 76, 8, 3374-3389, (2008).
  (Pubmed) (pdf)

Peer-Reviewed Conference Publications

• M.A. Kayala and P. Baldi.
  A machine learning approach to predict chemical reactions.
  In In J. Shawe-Taylor, R.S. Zemel, P. Bartlett, F.C.N. Pereira, and K.Q. Weinberger, editors, Advances in Neural Information Processing Systems 24, pages 747-755. (2011).
  (pdf)

Refereed Book Chapters

• J.H.Chen, M.A. Kayala, and P. Baldi.
  Reaction Explorer: Towards a knowledge map of organic chemistry to support dynamic assessment and personalized instruction.
  In R.E. Belford, J.W. Moore, and H.E. Pence, editors, Enhancing Learning with Online Resources, Social Networking, and Digital Libraries
  Volume 1060 of ACS Symposium Series. ACS, Washington D.C., (2010).
  (ACS) (pdf)

Theses

• M.A. Kayala.
  A comparison of Hensel lifting techniques for univariate polynomials.
  M.Sc. Thesis. University of Rhode Island, Department of Computer Science and Statistics, Kingston, RI. (2007).
  (pdf)

Presentations

• M.A. Kayala, C.A. Azencott, J.H. Chen, P. Baldi.
  Learning to predict chemical reactions.
  American Chemical Society National Meeting. Anaheim, CA. March, 2011.
  
  
• M.A. Kayala, C.A. Azencott, J.H. Chen, P. Baldi.
  OrbDB: Ranking Molecular Orbital Interactions.
  American Chemical Society National Meeting. San Francisco, CA. March, 2010.
  

Posters

• M.A. Kayala, C.A. Azencott, J.H. Chen, P. Baldi.
  A machine learning approach to predict chemical reactions.
  NLM Biomedical Informatics Annual Conference Bethesda, MD. June, 2011.
  Winner of the Best Poster Award.
  
  
• M.A. Kayala, C.A. Azencott, J.H. Chen, P. Baldi.
  Learning to predict chemical reactions.
  CINF Poster Session at the ACS National Meeting. Anaheim, CA. March, 2011.
  Winner of the Accelrys Scholarship for Scientific Excellence Award.
  
  
• M.A. Kayala, C.A. Azencott, J.H. Chen, P. Baldi.
  OrbDB: Ranking Molecular Orbital Interactions.
  CINF Poster Session at the ACS National Meeting. San Francisco, CA. March, 2010.

Teaching

• Spring 2012 - Reader
  CS 274b, Learning in Graphical Models (Graduate Class)
  University of California, Irvine
   
  
• Winter 2012 - Teaching Assistant
  ICS 21, Intro Comp Sci I
  University of California, Irvine
   
  
• Winter 2012 - Teaching Assistant
  ICS 22, Intro Comp Sci II
  University of California, Irvine
   
  
• Winter 2010 - Reader
  ICS 6D, Discrete Math
  University of California, Irvine
   
  
• 2005-2007 - Instructor
  CSC 200, Introduction to C++
  University of Rhode Island
   
  
• 2001 - Teaching Assistant
  PHY 204, Elementary Physics II
  University of Rhode Island
   
  
• 2000 - Teaching Assistant
  PHY 203, Elementary Physics I
  University of Rhode Island

Servers and Software

I have either developed or provided support for the following webservers.

• Machine learning based mechanistic reaction predictor:
  RxnPred

• Differential expression analysis:
  CyberT

• UCI Chemical Databank
  ChemDB

• Rule based chemical reactivity predictor. Used as pedagogical tool in Organic Chemistry classes.
  ReactionExplorer

I have made some of the underlying libraries for the above software available as open source at my Github.