Research highlights

I am strongly motivated by application, which serves as a base for developing new statistical methodologies. My goal is to create simple yet effective methods that are theoretically justified and useful to practitioners who need to draw scientific conclusions.

My current line of work involves improving on Bayesian non-negative matrix factorization algorithms, with a strong emphasis on cancer genomics and mutational signatures analysis.

During my Ph.D. at Duke, I developed Bayesian nonparametric methodologies under the supervision of Professor David Dunson within Project LIFEPLAN, an international research group whose aim is to quantify biodiversity across the globe.

I have been selected as a finalist for the 2024 Savage Award in Applied Methodology for my Ph.D. thesis. The winner will be announced August 2025 at JSM in Nashville. You can read my dissertation below.

Ph.D. Dissertation Ph.D. defense slides Google Scholar

Publications and manuscripts

Preprints

  1. Xue, C., Zito, A. and Miller, J. W. (2024+). Improved control of Dirichlet location and scale near the boundary. Invited Reject and Resubmit, working :) [ArXiv] [GitHub] - JSM 2025 SBSS student paper competition award received by Cathy Xue

  2. Zito, A. and Miller, J. W. (2024+). Compressive Bayesian non-negative matrix factorization for mutational signature analysis. Resubmitted after invited R&R. [ArXiv] [GitHub]

  3. Lee, C. J., Zito, A., Sang, H., and Dunson, D. B. (2024+). Logistic-Beta Processes for dependent random probabilities with beta marginals. Resubmitted after Major Revision. [ArXiv] [GitHub]


Articles in refereed journals

  1. Zito, A., Greaves, D., Soriano, J. and Richardson, L. (2025). Pareto optimal proxy metrics. Applied Stochastic Models in Business and Industry, 41: e70003 [Pub][ArXiv]

  2. Zito, A., Rigon, T., and Dunson, D. B. (2024). Bayesian nonparametric modeling of latent partitions via Stirling-gamma priors. Bayesian Analysis, [Pub][ArXiv] [R Package]

  3. Zito, A., Rigon, T., and Dunson, D. B. (2023). Inferring taxonomic placement from DNA barcoding aiding in discovery of new taxa. Methods in Ecology and Evolution, 14, 529–542. [Pub] [ArXiv] [Code] [R package]. - Best poster award at IBSA2022 world meeting.

  4. Zito, A., Rigon, T., Ovaskainen, O., and Dunson, D. B. (2023). Bayesian modeling of sequential discoveries. Journal of the American Statistical Association (T&M), 118(544), 2521–2532. [Pub] [Manuscript] [R package] - Best Student/Postdoc Contributed paper award at IBSA2021 world meeting. - BEST award for Ph.D. student research 2022, Dept. of Statistical Science, Duke University

Refereed conference proceedings and discussions

  1. Zito, A. and Miller, J. W. (2024). Discussion to “Sparse Bayesian factor analysis when the number of factors is unknown” by Fruhwirth–Schnatter, S., Hosszejini, D., and Lopes, F. L., Bayesian Analysis [Pub]

  2. Zito, A., Rigon, T. and Dunson, D. B. (2021). Modelling of accumulation curves through Weibull survival functions. In Book of Short Papers of the Italian Statistical Society 2021 (Perna, C., Salvati, N. and Schirripa Spagnolo, F., editors). ISBN: 9788891927361. [Link part 1] [Link part 2]


Manuscripts in preparation

  1. Zito, A., Xue, C., Parmigiani, G., and Miller, J. W. (2024+). Localized mutational signature analysis.

  2. Zito, A., Parmigiani, G., Miller, J. W. and Samur, M. (2024+). Paired Mutational signature analysis.

  3. Zito, A., Rigon T. and Dunson D. B. (2024+). Coviariate-based biodiversity measures via the Hubbell regression

  4. Orsholm, J.*, Zito, A.*, Chazlot, N, Roslin, T. and Furneaux, B. (2024+). Comparing model performance for taxonomic classification of unknown DNA barcodes.



Address:
655 Huntington Ave, Boston, MA 02115, United States
Email:
azito@hsph.harvard.edu
Links: