Edoardo (Edo) Airoldi

Associate Professor
Department of Statistics
FAS Center for Systems Biology
Harvard University
 
Principal Investigator
The Broad Insititute of MIT and Harvard

mailing address: Science Center, Harvard University, 1 Oxford Street, Cambridge, MA 02138, USA
phone: (617) 496-8318  
fax: (617) 496-8057
email: airoldi AT fas DOT harvard DOT edu

now.   My research interests encompass statistical methodology and theory with application to molecular biology and integrative genomics, computational social science, and statistical analysis of large biological and information networks. A few recent projects include (i) representation, sampling and inference methods for networks with rich node information, including the exchangeable graph model, (ii) metabolism and cellular proliferation, (iii) hypotheses testing and inference on signaling and metabolic pathways, including the map-kinase pathways, and the pathways regulating carbon and nitrogen metabolism, (iv) systems-level analyses of regulation and signaling dynamics with applications to stem cell development and cancer systems, (v) quantitation of alternative splicing events in human and mouse (vi) genome-wide quantitation of misreading and mistranslation events in yeast. Specific areas of technical interest include probabilitsic algorithms, approximation theorems, convex and combinatiorial optimization, and geometry.

brief bio.   I receved a Ph.D. from Carnegie Mellon in 2007, working on statistical machine learning and the analysis of complex systems with Stephen Fienberg and Kathleen Carley. My dissertation introduced statistical and computational elements of graph theory that support data analysis of complex systems and their evolution. Till December 2008, I was a postdoctoral fellow in the Lewis-Sigler Institute for Integrative Genomics of Princeton University working with Olga Troyanskaya, David Botstein, and James Broach. I developed mechanistic models to gain computational insights into aspects of the molecular and cellular biology that are not directly observable with experimental probes. I have been working closely with biologists and in the areas of cellular differentiation, cellular development and cancer, since.

  news   recent and upcoming talks
  • Graphlets: A scalable, multi-scale decomposition for large social and information networks. Stanford University (Algorithms for Modern Massive Data Sets), July 10-13, 2012, Palo Alto, CA.

  • Statistical analysis of populations with interacting and interfering units. Modern Modeling Methods Conference, May 23, 2012, Storrs, CT.

  • Valid inference in high-throughput biology and social media marketing. New England Machine Learning Day (Microsoft Research), May 16, 2012, Cambridge, MA.

  • Inference and algorithms for high-throughput biology. Microsoft Research, March 20, 2012, Redmond, WA.

  • Statistical methods for high-throughput biology. MIT (Bioinformatics at Math & CSAIL), February 22, 2012, Cambridge, MA.

  • Statistical analysis of populations with interacting and interfering units. Brown University (Applied Mathematics), February 1, 2012, Providence, RI.

  • Statistical analysis of populations with interacting and interfering units. Duke University (Statistics), January 20, 2012, Durham, NC.

  manuscripts
  • Bischof, JM. & Airoldi, EM. (2012).   Poisson convolution on a tree of categories.   (pdf) (a shorter version appeared at ICML 2012)

  • Costa, T., Guindani, M., Bassetti, F., Leisen, F. & Airoldi, EM. (2010).   Generalized species sampling priors with latent beta reinforcements.   (pdf)

  selected publications
  1. Airoldi, EM. & Blocker, AW. (2013).   Estimating latent processes on a network from indirect measurements.   Journal of the American Statistical Association.   (pdf)

  2. Choi, DS., Wolfe, PJ. & Airoldi, EM. (2012).   Stochastic blockmodels with growing number of classes.   Biometrika   (pdf)

  3. Slavov, N., Airoldi, EM., van Oudenaarden, A. & Botstein, D. (2012).   A conserved cell growth cycle can account for the environmental stress responses of divergent eukaryotes.   Molecular Biology of the Cell.   (pdf)

  4. Azari, H. & Airoldi, EM. (2011).   Graphlets decomposition of a weighted network.   Journal of Machine Learning Research, W&CP,   22 (AISTATS), 54-63. (pdf) (Recipient of the best student paper award, NESS 2012)

  5. Shieh, A., Hashimoto, T. & Airoldi, EM. (2011).   Tree preserving embedding.   Proceedings of the National Academy of Sciences,   108, 16916-16921. (pdf) (a shorter version appeared at ICML 2011)

  6. Airoldi, EM., Choi, DS. & Wolfe, PJ. (2011).   Confidence sets for network structure.   Statistical Analysis and Data Mining,   4, 461-469. (pdf) (a shorter version appeared at NIPS 2011)

  7. Blocker, AW & Airoldi, EM. (2011).   Deconvolution of mixing time series on a graph.   Uncertainty in Artificial Intelligence (UAI),   27, 51-60. (pdf) (Recipient of the IBM best student paper award, NESS 2011)

  8. Airoldi, EM & Haas, B. (2011).   Polytope samplers for inferece in ill-posed inverse problems.   Journal of Machine Learning Research, W&CP,   15 (AISTATS), 110-118. (pdf)

  9. Airoldi, EM., Heller, KA. & Silva, R. (2011).   Small sets of interacting proteins suggest functional linkage mechanisms via Bayesian analogical reasoning.   Bioinformatics,   27, i374-i382. (pdf)

  10. Airoldi, EM., Bai, X., & Carley, KM. (2011).   Network sampling and classification: An investigation of network model representations.   Decision Support Systems   51, 506-518. (pdf)

  11. Airoldi, EM., Bai, X., & Malin, B. (2011).   An entropy approach to disclosure risk assessment: Lessons from real applications and simulated domains.   Decision Support Systems,   51, 10-20. (pdf)

  12. Markowetz, F., Mulder, K.W., Airoldi, E.M., Lemischka, I., & Troyanskaya, O.G. (2010).   Mapping dynamic histone acetylation patterns to gene expression in Nanog-depleted murine embryonic stem cells.   PLoS Computational Biology,   6, e1001034. (pdf)

  13. Airoldi, EM., Erosheva, EA., Fienberg, SE., Joutard, CJ., Love, TM., & Shringarpure S. (2010).   Reconceptualizing the classification of PNAS articles.   Proceedings of the National Academy of Sciences,   107, 20899-20904. (pdf) (editorial)

  14. Katz, Y., Wang, E., Airoldi*, EM., & Burge*, CB. (2010).   Analysis and design of RNA sequencing experiments for identifying mRNA isoform regulation.   Nature Methods,   7, 1009-1015. (pdf) (supplement)

  15. Silva, R., Heller, K.A., Ghahramani, Z., & Airoldi E.M. (2010).   Ranking relations using analogies in biological and information networks.   Annals of Applied Statistics,   4, 615-644. (pdf)

  16. Goldenberg, A., Zheng, A.X., Fienberg, S.E., & Airoldi E.M. (2009).   A survey of statistical network models.   Foundations and Trends in Machine Learning,   2, 129-233. (pdf)

  17. Lu, R., Markowetz, F., Unwin, R.D., Leek, J.T., Airoldi E.M., MacArthur, B.D., Lachmann, A., Rozov, R., Ma’ayan, A., Boyer, L.A., Troyanskaya, O.G., Whetton, A.D., & Lemischka, I.R (2009).   Systems-level dynamic analyses of fate change in murine embryonic stem cells.   Nature,   462, 358-362. (pdf)

  18. Barutcuoglu, Z., Airoldi, E.M., Dumeaux, V., Schapire, R.E., & Troyanskaya, O.G. (2009).   Aneuploidy prediction and tumor classification with heterogeneous hidden conditional random fields.   Bioinformatics,   25, 1307-1313. (pdf)

  19. Airoldi, E.M., Huttenhower, C., Gresham, D., Lu, C., Caudy, A., Dunham, M., Broach, J., Botstein, D., & Troyanskaya, O.G. (2009).   Predicting cellular growth from gene expression signatures.   PLoS Computational Biology,   5, e1000257. (pdf)

  20. Airoldi, E.M., Blei, D.M., Fienberg, S.E., & Xing, E.P. (2008).   Mixed-membership stochastic blockmodels.   Journal of Machine Learning Research,   9, 1981-2014. (pdf) (a shorter version appeared at NIPS 2008)

  21. Warren, R., Airoldi, E.M., & Banks, D. (2008).   Network analysis of Wikipedia.   In: Statistical Methods in E-Commerce Research,   Jank & Shmueli, Eds., Wiley & Sons.

  22. Brauer, M.J., Huttenhower, C., Airoldi, E.M., Rosenstein, R., Matese, J.C., Gresham, D., Boer, V.M., Troyanskaya, O.G., & Botstein, D. (2008).   Coordination of growth rate, cell cycle, stress response and metabolic activity in yeast.   Molecular Biology of the Cell,   19, 352-367. (pdf)

  23. Airoldi, E.M. (2007).   Getting started in probabilistic graphical models.   PLoS Computational Biology,   3, e252. (pdf)

  24. Airoldi, E.M., Fienberg, S.E., Skinner, K.K. (2007).   Whose ideas? Whose words? Authorship of the Ronald Reagan radio addresses.   Political Science & Politics,   40, 501-506. (pdf)

  25. Airoldi, Blei, Fienberg, Goldenberg, Xing & Zheng, Eds. (2007).   Statistical Network Analysis: Models, Issues & New Directions.   Lecture Notes in Computer Science,   volume no. 4503. Springer-Verlag. (pdf)

  26. Airoldi, E.M. (2007).   Discussion on Handcock, Raftery and Tantrum 'Model-based clustering for social networks'.   Journal of the Royal Statistical Society, Series A,   170, Part 2, 30-31. (pdf)

  27. Malin, B., & Airoldi, E.M. (2007).   Confidentiality preserving audits of electronic medical record access.   World Congress on Health (Medical) Informatics (MEDINFO),   Brisbane, Australia. (pdf)

  28. Airoldi, E.M. (2006).   Bayesian mixed-membership models of complex and evolving networks.   Doctoral dissertation,   School of Computer Science, Carnegie Mellon University, PA USA. (pdf) (Recipient of the Savage award honorable mention, 2007)

  29. Airoldi, E.M., Anderson, A., Fienberg, S.E., Skinner, K.K. (2006).   Who wrote Ronald Reagan's radio addresses?   Bayesian Analysis,   1, 289-320. (pdf) (tr stat-03-789) (notes)

  30. Airoldi, E.M., & Carley, K.M. (2005).   Sampling algorithms for pure network topologies: Stability and separability of metric embeddings.   ACM KDD Explorations (Special Issue on Link Mining),   7, 13-22.

  31. Airoldi, E.M., Blei, D.M., Fienberg, S.E., & Xing, E.P. (2005).   A latent mixed-membership model for relational data.   ACM KDD Workshop on Link Analysis,   Chicago IL. (pdf) (Recipient of the John Van Ryzin award, 2006)

  32. Airoldi, E.M., Cohen, W.W., & Fienberg, S.E. (2005).   Bayesian methods for frequent terms in text: Models of contagion and the Delta square statistic.   CSNA & INTERFACE Annual Meetings,   St. Louis MI. (pdf)

  33. Malin, B., Airoldi, E.M., & Carley, K.M. (2005).   A social network analysis model for name disambiguation in lists.   Journal of Computational and Mathematical Organization Theory,   11, 119-139. (pdf)

  34. Gross, R., Airoldi, E.M., Malin, B., & Sweeney, L.A. (2005).   Integrating utility into face de-identification.   Privacy Enhancing Technologies (Revised Selected Papers),   Lecture Notes in Computer Science, vol. 3856, 227-242. (pdf)

  35. Malin, B., Airoldi, E.M., Edoho-Eket, S., & Li, Y. (2005).   Configurable security protocols for multi-party data analysis with malicious participants.   IEEE International Conference onf Data Engineering (ICDE),   Tokyo Japan. (pdf)

  36. Airoldi, E.M., Malin, B., & Sweeney, L.A. (2005).   Technologies to defeat fraudulent schemes related to email requests.   AAAI Symposium on Artificial Intelligence Technologies for Homeland Security,   Stanford CA. (pdf)

  37. Airoldi, E.M., & Faluotsos, C. (2004).   Recovering latent time series from their observed sums: Network tomography with particle filters.   ACM International Conference on Knowledge Discovery & Data Mining (KDD),   Seattle WA. (pdf) (tr cs-04-130) (Runner-up for best paper award)

  38. Airoldi, E.M. (2000).   Forecasts for Italian and Greek macroeconomic indicators.   Global Data Watch,   J.P. Morgan Economic Research. (issues of March 6-13-20-27, April 3-10-17-24, May 1-8)

  39. Fornari, I., & Airoldi, E.M. (1999).   Assessing Italy's inflation gap relative to the Euro area.   Global Data Watch,   J.P. Morgan Economic Research. (issue of December 20)

  40. Airoldi, E.M. (1999).   The theory of weak convergence of probability measures and its applications in statistics.   Dissertation,   Institute for Quantitative Methods, Bocconi University, Milan Italy. (Gold medal for best graduates, 1999)

  collaborators
  • David Banks (Duke), David Blei (Princeton), David Botstein (Princeton), Chris Burge (MIT), Kathleen Carley (Carnegie Mellon), Stephen Fienberg (Carnegie Mellon), Katherine Heller (University College London), Curtis Huttenhower (Princeton), Jure Leskovec (Carnegie Mellon), Xiodong Lin (University of Cincinnati), Bradley Malin (Vanderbilt), Florian Markowetz (Princeton), Erin O'Shea (Harvard), Aviv Regev (Broad Institute), Ricardo Silva (University College London), Olga Troyanskaya (Princeton), Eric Xing (Carnegie Mellon).