All publications with annotations and links to talks

 

·         T. Bahadori, E. T. Tchetgen, D. Heckerman.  End-to-end balancing for causal continuous treatment-effect estimation.  ICML, July 2022.

 

We bring together methods from causal inference and machine learning to improve the accuracy of propensity score weighting.

 

·         R. Shachter, D. Heckerman.  Why did they do that? In Probabilistic and Causal Inference: The Works of Judea Pearl. Association for Computing Machinery, New York, NY. Feb 2022.

 

We argue that decision-theoretic thinking, which includes the ability to identify possible actions, the ability to imagine the outcomes of those actions (i.e., causal reasoning), and the ability to select among the outcomes offers strong survival advantages. We note that such thinking is modular and thus computationally feasible, which suggests why it may have evolved in humans. Finally, we note that the perception of free will facilitates such thinking and ask whether this perception is a necessary component of decision-theoretic thinking.

 

·         T. Bahadori, D. Heckerman.  Debiasing concept-based explanations with causal analysis. The Tenth International Conference on Learning Representations, April 2021.

 

We bring causal analysis to the problem concept-based explanation in machine learning. We show how two-stage regression used in causal inference can be used to improve explanations.

 

·         G.M. Souza, M.A. Van Sluys, C.G. Lembke, H. Lee, G.R.A. Margarido, C.T. Hotta, J.W. Gaiarsa, A.L. Diniz, M. de Medeiros Oliveira, S. de Siqueira Ferreira, M.Y. Nishiyama Jr, F. ten-Caten, G.T. Ragagnin, P. de Morais Andrade, R.F. de Souza, G.G. Nicastro, R. Pandya, C. Kim, H. Guo, A.M. Durham, M. S. Carneiro, J. Zhang, X. Zhang, Q. Zhang, R. Ming, M.C.Schatz, R. Davidson, A.H. Paterson, and D. Heckerman.  Assembly of the 373k gene space of the polyploid sugarcane genome reveals reservoirs of functional diversity in the world's leading biomass crop.  GigaScience, 8(12), Dec 2019 (doi.org/10.1093/gigascience/giz129).

 

·         A. Wells, D. Heckerman, A. Torkamani, L. Yin, J. Sebat, B. Ren, A. Telenti, and  J. di Iulio.  Ranking of non-coding pathogenic variants and putative essential regions of the human genome.  Nature Communications, 10, Nov, 2019 (doi: 10.1038/s41467-019-13212-3).

 

·         D. Gurdasani, T. Carstensen, S. Fatumo, G. Chen, C.S. Franklin, J. Prado-Martinez, H. Bouman, F. Abascal, M. Haber, I. Tachmazidou, I. Mathieson, K. Ekoru, M.K. DeGorter, R.N. Nsubuga, C. Finan, E. Wheeler, L. Chen, D.N. Cooper, S. Schiffels, Y. Chen, G.R.S. Ritchie, M.O. Pollard, M.D. Fortune, A.J. Mentzer, E. Garrison, A. Bergström, K. Hatzikotoulas, A. Adeyemo, A. Doumatey, H. Elding, L.V. Wain, G. Ehret, P.L. Auer, C.L. Kooperberg, A.P. Reiner, N. Franceschini, D.P. Maher, S.B. Montgomery, C. Kadie, C. Widmer, Y. Xue, J. Seeley, G. Asiki, A. Kamali, E. H. Young, C. Pomilla, N. Soranzo, E. Zeggini, F. Pirie, A.P. Morris, D. Heckerman, C. Tyler-Smith, A. Motala, C. Rotimi, P. Kaleebu, I. Barroso, M.S. Sandhu.  Uganda genome resource enables insights into population history and genomic discovery in Africa.  Cell, 179(4): 984-1002, Oct 2019 (doi: 10.1016/j.cell.2019.10.004).

 

·         D. Heckerman.  Toward accounting for hidden common causes when inferring cause and effect from observational data.  ACM Transactions on Intelligent Systems and Technology, 10, Sept 2019 (doi: 10.1145/3309720).  Preprint Jan 3, 2018.  Talk at KDD 2019.

 

·         S. Lee, N. Gornitz, E.P. Xing, D. Heckerman, and C. Lippert.  Ensembles of Lasso Screening Rules.  IEEE Transactions on Pattern Analysis and Machine Intelligence, 40, Dec 2018.

 

·         J. Zhang, X. Zhang, H. Tang, Q. Zhang, X. Hua, X. Ma, F. Zhu, T. Jones, X. Zhu, J. Bowers, C.7 Wai, C. Zheng, Y. Shi, S. Chen, X. Xu, J. Yue, D.R. Nelson, L. Huang, Z. Li, H. Xu, D. Zhou, Y. Wang, W. Hu, J. Lin, Y. Deng, N. Pandey, M. Mancini, D. Zerpa, J.K. Nguyen, L. Wang, L. Yu, Y. Xin, L. Ge, J. Arro, J.O. Han, S. Chakrabarty, M. Pushko, W. Zhang, Y. Ma, P. Ma, M. Lv, F. Chen, G. Zheng, J. Xu, Z. Yang, F. Deng, X. Chen, Z. Liao, X. Zhang, Z. Lin, H. Lin, H. Yan, Z. Kuang, W. Zhong, P. Liang, G. Wang, Y. Yuan, J. Shi, J. Hou, J. Lin, J. Jin, P. Cao, Q. Shen, Q. Jiang, P. Zhou, Y. Ma, X. Zhang, R. Xu, J. Liu, Y. Zhou, H. Jia, Q. Ma,  R. Qi, Z. Zhang, J. Fang, H. Fang, J. Song, M. Wang, G. Dong, G. Wang, Z. Chen, T. Ma, H. Liu, S.R. Dhungana, S.E. Huss, X. Yang, A. Sharma, M.C. Martinez, M. Hudson, J.J. Riascos, M. Schuler, L.Q. Chen, D.M. Braun L. Li, Q. Yu, J. Wang, K. Wang, M.C. Schatz, D. Heckerman, M.A. Van Sluys, G.M. Souza, P.H. Moore, D. Sankoff, R. VanBuren, A. H. Paterson, C. Nagai, R. Ming.  Allele-defined genome of the autopolyploid sugarcane Saccharum spontaneum L.  Nature Genetics, 50: 1565-1573, Nov 2018 (doi: 10.1038/s41588-018-0237-2).

 

·         C. Kadie, D. Heckerman.  Ludicrous Speed Linear Mixed Models for Genome-Wide Association Studies. BioRXiv, Jan 2018.

 

·         H. Tang, E. Kirkness, C. Lippert, W. Biggs, M. Fabani, E. Guzman, S. Ramakrishnan, V. Lavrenko, B. Kakaradov, C. Hou, B. Hicks, D. Heckerman, F.J. Och, C.T. Caskey, J.C. Venter, A. Telenti.  Profiling of Short-Tandem-Repeat Disease Alleles in 12,632 Human Whole Genomes.  AJHG, 101, 700-715, Nov 2017 (doi: 10.1016/j.ajhg.2017.09.013).

 

·         D. Heckerman, B. Traynor, A. Picca, R. Calvani, E. Marzetti, D. Hernandez, M. Nalls, S. Arepali, L. Ferrucci, and F. Landi.  Genetic variants associated with physical performance and anthropometry in old age: A genome-wide association study in the ilSIRENTE cohort.  Scientific Reports 7, 15879, Nov 2017 (doi:10.1038/s41598-017-13475-0).

 

·         D. Heckerman, D. Gurdasani, C. Kadie, C. Pomilla, T. Carstensen, H. Martin, K. Ekoru, R.N. Nsubuga, G. Ssenyomo A. Kamali, P. Kaleebu, C. Widmer, and M.S. Sandhu.  Linear mixed model for heritability estimation that explicitly addresses environmental variation.  PNAS, 113: 7377–7382, July 2016 (doi: 10.1073/pnas.1510497113).  Talk at ASHG 2015.

 

Describes a way to generalize linear mixed models to take spatial location into account when jointly modeling the influences of genomics and environment on traits.  The high-level message from this work is that estimates of heritability are extremely sensitive to model specification.  Therefore, as it is usually difficult to guarantee a good specification, such estimates should be treated with skepticism.  Erratum: After Equation 3, the sentence should read “Note that element i,j of Kcausal is proportional to the dot product….”

 

·        J.M. Carlson, V.Y. Du, N. Pfeifer, A. Bansal, V.Y.F. Tan, K. Power, C.J. Brumme, A. Kreimer, C.E. DeZiel, N. Fusi, M. Schaefer, M.A. Brockman, J. Gilmour, M.A. Price, W. Kilembe, R. Haubrich, M. John, S. Mallal, R. Shapiro, J. Frater, P.R. Harrigan, T. Ndung'u, S. Allen, D. Heckerman, T.M. Allen, P.J.R. Goulder, Z.L. Brumme, E. Hunter & P.A. Goepfert.  Impact of pre-adapted HIV transmission.  Nature Medicine 22, 606¡V613 (2016) (doi:10.1038/nm.4100).

 

·        C. Lippert and D. Heckerman.  Computational and statistical issues in personalized medicine.  XRDS 21, 24-27, Summer 2015 (doi:10.1145/2788502).

 

      Describes statistical issues in GWAS with linear mixed models from a graphical-model perspective.

 

·         K. Pfafferott, P. Deshpande, E. McKinnon, S. Merani, A. Lucas, S. Mallal, D. Heckerman, M. John, S. Gaudieri, M. Lucas.  Anti-Hepatitis C virus T-cell immunity in the context of multiple exposures to the virus.  PLoS One, June 2015 (doi:10.1371/journal.pone.0130420).

 

·        C. Berger, A. Llano, J. Carlson, Z. Brumme, Mark Brockman, S. Cedeño, P.R. Harrigan, D. Kaufmann, D. Heckerman, A. Meyerhans, and C. Brander.  Immune screening identifies novel T cell targets encoded by antisense reading frames of HIV-1.  J. Virol 89, 4015-4019, April 2015 (doi:10.1128/JVI.03435-14).

 

·        G. Margarido and D. Heckerman. ConPADE: Genome Assembly Ploidy Estimation from Next-Generation Sequencing Data. PLoS Comput Biol 11(4): e1004229, April 2015 (doi:10.1371/journal.pcbi.1004229).

 

·        O. Weissbrod, C. Lippert, D. Geiger, and D. Heckerman.  Accurate liability estimation improves power in ascertained case-control studiesNature Methods, Feb 2015 (doi:10.1038/nmeth.3285).  Preprint.

 

      Describes an approach to pre-process ascertained case-control-study data that leads to improved power when analyzed with a linear mixed model.

 

·        C. Berger, A. Llano, J. Carlson, Z. Brumme, M. Brockman, S. Cedeño, P. Harrigan, D. Kaufmann, D. Heckerman, A. Meyerhans, C. Brander.  Immune screening identifies novel T cell targets encoded by anti-sense reading frames of HIV-1J Virol, Jan 2015.

 

·        C. Widmer, C. Lippert, O. Weissbrod, N. Fusi, C.M. Kadie, R.I. Davidson, J. Listgarten, and D. Heckerman. Further Improvements to Linear Mixed Models for Genome-Wide Association Studies. Scientific Reports 4, 6874, Nov 2014 (doi:10.1038/srep06874).  Talk at the Sackler Big Data Colloquium, Washington DC 2015.

 

Describes the latest version of FaST-LMM.  It shows that selecting SNPs for the linear-mixed-model similarity matrix through pruning via linkage disequilibrium (e.g., selecting every kth SNP) works well to control type I error, whereas selecting SNPs that are predictive of the phenotype does not.

 

·        F. Pereyra, D. Heckerman, J. Carlson, C. Kadie, D. Soghoian, D. Karel, A. Goldenthal, O. Davis, C. DeZiel, T. Lin, J. Peng, A. Piechocka, M. Carrington, and B. Walker. HIV Control Is Mediated in Part by CD8+ T-Cell Targeting of Specific Epitopes. J. Virol 88 12937-12948, Aug 2014.  A talk on the history of junk-mail filtering and its relationship to this work was given at the Microsoft Worldwide Partner Conference 2013.  Talk at CROI 2009 on an early version of this work.

 

      Describes the design of an HIV vaccine based on the idea that particular regions of the HIV genome are more vulnerable to attack by the cellular immune system than other regions.

 

·        J. Kuipers, G. Moffa, and D. Heckerman. Addendum on the scoring of Gaussian directed acyclic graphical models. Annals of Statistics 42, 1689-1691, Aug 2014.

 

·        D. Heckerman, C. Meek, and T. Richardson. Variations on undirected graphical models and their relationships. Kybernetika 50, 363–377, July 2014.  Microsoft copy.  Original 2004 version.

       

·        C. Lippert, J. Xiang, D. Horta, C. Widmer, C.M. Kadie, D. Heckerman, J. Listgarten. Greater Power and Computational Efficiency for Kernel-Based Association Testing of Sets of Genetic Variants. Bioinformatics 30 , July 2014 (doi: 10.1093/bioinformatics/btu504).

 

Shows empirically and provides theoretical arguments that the LRT can be more powerful than a score test for set association tests.  Also describes how to do a number of algebraic computations efficiently for set tests with either a low- or full-rank background kernel.

 

·        J. Carlson, M. Schaefer, D. Monaco, R. Batorsky, D. Claiborne, J. Prince, M. Deymier, Z. Ende, N. Klatt, C. DeZiel, Tien-Ho Lin, J. Peng, A. Seese, R. Shapiro, J. Frater, T. Ndung’u, J. Tang, P. Goepfert, J. Gilmour, M. Price, W. Kilembe, D. Heckerman, P. Goulder, T. Allen, S. Allen, and E. Hunter. Selection bias at the heterosexual HIV-1 transmission bottleneck.  Science, 345(6193): 1254031, July 2014 (doi:10.1126/science.1254031).

     

Provides evidence that a cellular HIV vaccine could be sterilizing.

 

·        H. Poon, C. Quirk, C. DeZiel, and D. Heckerman. Literome: PubMed-scale genomic knowledge base in the cloud. Bioinformatics 30, 2840-2842, June 2014.

 

·        R. Rubsamen, C. Herst, P. Lloyd, D. Heckerman. Eliciting cytotoxic T-lymphocyte responses from synthetic vectors containing one or two epitopes in a C57BL/6 mouse model using peptide-containing biodegradable microspheres and adjuvants. Vaccine 32, 4111-4116, June 2014.

 

      Describes a delivery mechanism for the HIV vaccine design described in Pereyra, et al., J. Virol, 2014.

 

·       N. Furlotte, D. Heckerman, and C. Lippert.  Quantifying the uncertainty in heritabilityJournal of Human Genetics 27, March 2014 (doi: 10.1038/jhg.2014.15).

 

Applies the spectral-decomposition trick from FaST-LMM (Lippert et al., Nature Methods, 2011) to speed up Bayesian estimates of heritability.

 

·        J. Zou, C. Lippert, D. Heckerman, M. Aryee, and J. Listgarten.  Epigenome-wide association studies without the need for cell-type compositionNature Methods 11:309–311, Jan 2014 (doi:10.1038/nmeth.2815).

 

Applies FaST-LMM (Lippert et al., Nature Methods, 2011) to the analysis of epigenetic data.  Principal components are added to correct for the confounding effects of multiple cell types. 

Regarding the identification of principal components, the approach in this paper is inferior to the one in Widmer et al., Scientific Reports, 2014.

 

·        J. Sunshine, M. Kim, J. Carlson, D. Heckerman, J. Czartoski, S. Migueles, J. Maenza, M. McElrath, J. Mullins, and N. Frahm. Increased sequence coverage through combined targeting of variant and conserved epitopes correlates with control of HIV replicationJournal of Virology 88(2):1354-65, Jan 2014 (doi: 10.1128/JVI.02361-13).

 

·        I. Bartha, J.Carlson, C. Brumme, P. McLaren, Z. Brumme, M. John, D. Haas, J. Martinez-Picado, J. Dalmau, C. López-Galíndez, C. Casado, A. Rauch, H. Günthard, E. Bernasconi, P. Vernazza, T. Klimkait, S. Yerly, S. O’Brien, J. Listgarten, N. Pfeifer, C. Lippert, N. Fusi, Z. Kutalik, T. Allen, V. Müller, P. Harrigan, D. Heckerman, A. Telenti, and J. Fellay.  A genome-to-genome analysis of associations between human genetic variation, HIV-1 sequence diversity, and viral controleLife 2013;2:e01123, October 2013 (http://dx.doi.org/10.7554/eLife.01123).

 

·        C. Lippert, G. Quon, E.Y. Kang, C.M. Kadie, J. Listgarten, and D. Heckerman.  The benefits of selecting phenotype-specific variants for applications of mixed models in genomicsScientific Reports, 3, May 2013 (doi:10.1038/srep01815).

 

This work describes a method for selecting SNPs for the linear-mixed-model similarity matrix by identifying SNPs that are predictive of the phenotype.  A later publication Widmer et al., Scientific Reports, 2014 shows that selecting phenotype-predictive variants yields poor control of type I error, and shows that selecting SNPs through pruning via linkage disequilibrium (e.g., selecting every kth SNP) works better.  The latter approach is recommended. 

 

On the positive side, this work corrects an error in Listgarten et al., Nature Methods, 2012 regarding an explanation for “dilution,” the phenomenon in which the inclusion of irrelevant SNPs in the similarity matrix leads to inflation of the test statistic lambda and reduced power.

 

·        J. Listgarten, C. Lippert, E.Y. Kang, J. Xiang, C.M. Kadie, and D. Heckerman.  A powerful and efficient set test for genetic markers that handles confoundersBioinformatics, May 2013 (doi: 10.1093/bioinformatics/btt177).

 

Shows that the likelihood-ratio test can be more powerful than a score test for set association tests.  This work is limited to similarity matrices that are low rank and includes an efficient algorithm for this case.  (See Lippert et al., Bioinformatics 2014 for extensions.)

 

·        J. Listgarten, C. Lippert, and D. Heckerman. FaST-LMM-Select for addressing confounding from spatial structure and rare variantsNature Genetics, 45: 470-471, April 2013 (doi:10.1038/ng.2620).

 

Shows how using SNPs in a linear-mixed-model similarity matrix that are predictive of the phenotype solves an open problem in statistical genetics that had been published in Nature Genetics. A later publication Widmer et al., Scientific Reports, 2014 shows this approach yields poor control of type I error in many circumstances, and shows that selecting SNPs through pruning via linkage disequilibrium (e.g., selecting every kth SNP) works better.  The latter approach is recommended.

 

·        C. Boutwell, J.M. Carlson, T. Lin, A. Seese, K. Power, J. Peng, Y. Tang, Z. Brumme, D. Heckerman, A. Schneidewind, and T. Allen.  Frequent and variable cytotoxic-T-lymphocyte escape-associated fitness costs in the human immunodeficiency virus type 1 subtype B Gag proteinsJournal of Virology, 87(7): 3952-3965, April 2013 (10.1128/JVI.03233-12).

 

·        G. Quon, C. Lippert, D. Heckerman, and J. Listgarten.  Patterns of methylation heritability in a genome-wide analysis of four brain regionsNucleic Acids Research, March 2013 (doi: 10.1093/nar/gks1449).

 

·        C. Bronke, C.A. Almeida, E. McKinnon, S.G. Roberts, N.M.  Keane, A. Chopra, J.M. Carlson, D. Heckerman, S. Mallal, and M. John.  HIV escape mutations occur preferentially at HLA-binding sites of CD8 T-cell epitopesAIDS, 27(6): 899–905, March 2013 (doi: 10.1097/QAD.0b013e32835e1616).

 

·        V. Kulkarni, M. Rosati, A. Valentin, B. Ganneru, A.K. Singh, J. Yan, M. Rolland, C. Alicea, R.K. Beach, G. Zhang, S. Le Gall, K.E. Broderick, N.Y. Sardesai, D. Heckerman, B. Mothe, C. Brander, D.B. Weiner, J.I. Mullins, G.N. Pavlakis, B.K. Felber.  HIV-1 p24-Gag derived conserved element DNA vaccine increases the breadth of immune response in micePLoS ONE 8(3): e60245, March 2013 (doi:10.1371/journal.pone.0060245).

 

·       C. Lippert, J. Listgarten, R.I. Davidson, J. Baxter, H. Poon, C.M. Kadie, and  D. Heckerman.  An Exhaustive Epistatic SNP Association Analysis on Expanded Wellcome Trust DataScientific Reports, 3, Jan 2013 (doi:10.1038/srep01099).

 

Presents results for all pairwise-epistatic tests for all phenotypes in the WTCCC1 data, using a linear mixed model.  The approach improves power by expanding the control set to include other disease cohorts, multiple races, and closely related individuals.  One caveat is that the approach uses a low-rank similarity matrix based on SNPs that are predictive of the phenotype. A later publication Widmer et al., Scientific Reports, 2014 shows that this method for selecting SNPs yields poor control of type I error, and shows that selecting SNPs through pruning via linkage disequilibrium (e.g., selecting every kth SNP) works better.  The latter approach is recommended.

 

·        Z.M. Ndhlovu, L.B. Chibnik, J. Proudfoot, S. Vine, A. McMullen, K. Cesa, F. Porichis, R.B. Jones, D.M. Alvino, M.G. Hart, E. Stampouloglou, A. Piechocka-Trocha, C.M. Kadie, F. Pereyra, D. Heckerman, P.L. De Jager, B.D. Walker, and D.E. Kaufmann.  High-dimensional immunomonitoring models of HIV-1–specific CD8 T-cell responses accurately identify subjects achieving spontaneous viral control.  Blood, 121(5): 801-811, Jan 2013 (doi: 10.1182/blood-2012-06-436295).

 

·        J. Carlson, C. Brumme, E. Martin, J. Listgarten, M. Brockman, A. Le, C. Chui, L. Cotton, D. Knapp, S. Riddler, R. Haubrich, G. Nelson, N. Pfeifer, C. DeZiel, D. Heckerman, R. Apps, M. Carrington, S. Mallal, R. Harrigan, M. John, and Z. Brumme.  Correlates of protective cellular immunity revealed by analysis of population-level immune escape pathways in HIV-1.  Journal of Virology, 86(24): 13202-13216, Dec 2012 (doi: 10.1128/JVI.01998-12).

 

·        P. C. Matthews, M. Koyanagi, H. N. Kloverpris, M. Harndahl, A. Stryhn, T. Akahoshi, H. Gatanaga  S. Oka, C. Juarez Molina, P. H. Valenzuela, R. Avila Rios, D. Cole, J. Carlson, R. P. Payne, A. Ogwu, A. Bere, T. Ndung'u, K. Gounder, F. Chen, L. Riddell, G. Luzzi, R. Shapiro, C. Brander, B. Walker, A. K. Sewell, G. Reyes Teran, D. Heckerman, E. Hunter, S. Buus, M. Takiguchi, P. J. GoulderJ.  Differential clade-specific HLA-B*3501 association with HIV-1 disease outcome is linked to immunogenicity of a single Gag epitopeJournal of Virololgy, 86(23): 12643-54, Dec 2012 (doi: 10.1128/JVI.01381-12).

 

·        R. Apps, Y. Qi, J. M. Carlson, H. Chen, X. Gao, R. Thomas, Y. Yuki, G. Q. Del Prete, P. Goulder, Z. L. Brumme, C. J. Brumme, M John, S. Mallal, G. Nelson, R. Bosch, D. Heckerman, J. L. Stein, K. A. Soderberg, M. A. Moody, T. N. Denny, Xue Zeng, J. Fang, A. Moffett, J. D. Lifson, J. J. Goedert, S. Buchbinder, G. D. Kirk, J. Fellay, P. McLaren, S. G. Deeks, F. Pereyra, B. Walker, N. L. Michael, A. Weintrob, S. Wolinsky, W. Liao, M. Carrington.  Influence of HLA-C Expression Level on HIV Control.  Science, 340(6128): 87-91, Nov 2012 (doi: 10.1128/JVI.02122-12).

 

·        J.L. Prince, D.T. Claiborne, J.M. Carlson, M. Schaefer, T. Yu, S. Lahki, H.A. Prentice, L.Yue, S.A. Vishwanathan, W. Kilembe, P. Goepfert, M.A. Price, J. Gilmour, J. Mulenga, P. Farmer, C.A. Derdeyn, J. Tang, D. Heckerman, R.A. Kaslow, and S.A. Allen.  Role of transmitted Gag CTL polymorphisms in defining replicative capacity and early HIV-1 pathogenesisPLoS Pathogens 8(11): e1003041, Nov 2012 (doi:10.1371/journal.ppat.1003041).

 

·        S. Nomura, N. Hosoya, Z. Brumme, M. Brockman, T. Kikuchi, M. Koga, H. Nakamura, T. Koibuchi, T. Fujii, J. Carlson, D. Heckerman, A. Tachikawa, A. Iwamoto, and T. Miura.  Significant reductions in Gag-protease mediated HIV-1 replication capacity over the course of the epidemic in JapanJournal of Virology, 87(3): 1465-1476, Nov 2012 (doi: 10.1128/JVI.02122-12).

 

·        P.C. Matthews, J. Listgarten, J.M. Carlson, R. Payne, K.G. Huang, J. Frater, D. Goedhals, D. Steyn, C. van Vuuren, P. Paioni, P. Jooste, A. Ogwu, R. Shapiro, Z. Mncube, T. Ndung'u, B.D. Walker, D. Heckerman, and P.J.R. Goulder.  Co-operative additive effects between HLA alleles in control of HIV-1.  PLoS One 7(10): e47799, Oct 2012 (doi:10.1371/journal.pone.0047799).

 

·        D. Jacobson, A. Parker, C. Spetzler, W. de Bruin, K. Hollenbeck, D. Heckerman, B. Fischhoff.  Improved Learning in U.S. History and Decision Competence with Decision-Focused Curriculum.  PLoS One, Sept 2012 (doi:10.1371/journal.pone.0045775).

 

·        J. Listgarten, C. Lippert, C.M. Kadie, R.I. Davidson, E. Eskin, and D. Heckerman.  Improved linear mixed models for genome-wide association studiesNature Methods, 9: 525-526, June 2012 (doi:10.1038/nmeth.2037).

 

Describes a method for selecting SNPs for the linear-mixed-model similarity matrix by identifying SNPs that are predictive of the phenotype.  A later publication Widmer et al., Scientific Reports, 2014 shows this approach yields poor control of type I error, and shows that selecting SNPs through pruning via linkage disequilibrium (e.g., selecting every kth SNP) works better.  This work also shows that the inclusion of irrelevant SNPs in the similarity matrix leads to inflated test statistic lambda and reduced power, a phenomenon called “dilution”.  Although an incorrect explanation for dilution is offered here, a correction is given in Lippert et al., Scientific Reports, 2013.  Finally, there is a bug in the analysis of the synthetic data (Figure S1), which makes the prediction-based selection method appear to perform better than it actually does.

 

·        P.J. McLaren, S. Ripke, K. Pelak, A.C. Weintrob, N.A. Patsopoulos, X. Jia, R.L. Erlich, N.J. Lennon, C.M. Kadie, D. Heckerman, N. Gupta, D.W. Haas, S.G. Deeks, F. Pereyra, B.D. Walker, P.I.W. de Bakker, and the International HIV Controllers Study.  Fine-mapping classical HLA variation associated with durable host control of HIV-1 infection in African Americans.  Human Molecular Genetics, 21(19): 4334-4347, June 2012 (doi: 10.1093/hmg/dds226).

 

·        X. Zhang, W. Cheng, J. Listgarten, C. Kadie, S. Huang, W. Wang, and D. Heckerman.  Learning transcriptional regulatory relationships using sparse graphical modelsPLoS One 7(5): e35762, May 2012 (doi:10.1371/journal.pone.0035762).

 

·        M.Brockman, D. Chopera, A. Olvera, C. Brumme, J. Sela, T. Markle, E Martin, J.Carlson, A. Le, R. McGovern, P. Cheung, A. Kelleher, H. Jessen, M. Markowitz, E. Rosenberg, N. Frahm, J. Sanchez, S. Mallal, M. John, P. Harrigan, D. Heckerman, C. Brander, B. Walker, and Z. Brumme.  Uncommon pathways of immune escape attenuate HIV-1 integrase replication capacityJournal of Virology, April, 2012 (doi: 10.1128/JVI.07133-11).

 

·        J. Carlson, J. Listgarten, N. Pfeifer, V. Tan, C. Kadie, B. Walker, T. Ndung'u, R. Shapiro, J. Frater, Z. Brumme, P. Goulder, and D. Heckerman.  Widespread impact of HLA restriction on immune control and escape pathways in HIV-1.BJournal of Virology, February, 2012 (doi: 10.1128/JVI.06728-11).

 

·        B. Mothe, A. Llano, J. Ibarrondo, J. Zamarre, M. Schiaulini, C. Miranda, M. Ruiz-Riol, C. Berger, M. J. Herrero, E. Palou, M. Plana, M. Rolland, A. Khatri, D. Heckerman, F. Pereyra, B. Walker, D. Weiner, R. Paredes, B. Clotet, B. Felber, G. Pavlakis, J. Mullins, C. Brander.  CTL Responses of High Functional Avidity and Broad Variant Cross-Reactivity are Associated with HIV ControlPLoS ONE 7(1):  e29717, Jan 2012 (doi:10.1371/journal.pone.0029717).

 

·        B. Mothe, A. Llano, J. Ibarrondo, M. Daniels, C. Miranda, J. Zamarreno, V. Bach, R. Zuniga, S. Perez-Alvarez, C. Berger, M. Puertas, J. Martinez-Picado, M. Rolland, M. Farfan, J. Szinger, W. Hildebrand, O. Yang, V. Sanchez-Merino, C. Brumme, Z. Brumme, D. Heckerman, T. Allen, J. Mullins, G. Gomez, P. Goulder, B.Walker, J. Gatell, B. Clotet, B. Korber, J. Sanchez and C. Brander.  Definition of the viral targets of protective HIV-1-specific T cell responsesJournal of Translational Medicine, 9: 208, Dec 2011 (doi:10.1186/1479-5876-9-208).

 

·        A. Renton, et al.  A Hexanucleotide Repeat Expansion in C9ORF72 Is the Cause of Chromosome 9p21-Linked ALS-FTDNeuron 72(2): 257-268, Oct 2011.

 

·        S. Ranasinghe, M. Flanders, S. Cutler, D. Soghoian, M. Ghebremichael, I. Davis, M. Lindqvist, F. Pereyra, B. Walker, D. Heckerman, and H. Streeck.  HIV-specific CD4 T cell responses to different viral proteins have discordant associations with viral load and clinical outcomeJournal of Virology, Oct 2011 (doi: 10.1128/JVI.05577-11).

 

·        C. Lippert, J. Listgarten, Y. Liu, C.M. Kadie, R.I. Davidson, and D. Heckerman.  FaST linear mixed models for genome-wide association studiesNature Methods, 8: 833-835, Oct 2011 (doi:10.1038/nmeth.1681).  Preprint.  Talk at the Big Data for Precision Health Conference, Stanford 2015.

 

Shows how exact linear-mixed-model computations can be performed in time and memory linear in the number of individuals when the number of SNPs used in the similarity matrix is less than the number of individuals (i.e., when the similarity matrix is low rank).  In addition, this work shows that exact computations are quadratic in time and memory when the similarity matrix is full rank. This work also describes an approach to select SNPs to achieve this condition with linkage-disequilibrium-based pruning (e.g., selecting every kth SNP).

 

·        G. Alter, D. Heckerman, A. Schneidewind, L. Fadda, C. Kadie, J. Carlson, C. Oniangue-Ndza, M. Martin, B. Li, S. Khakoo, M. Carrington, T. Allen, M. and Altfeld M.  HIV-1 adaptation to NK-cell-mediated immune pressureNature, 476 (7358): 96-100, August 2011.

 

·        C. Almeida, C. Bronke, S. Roberts, E. McKinnon, N. Keane, A. Chopra, C. Kadie, J. Carlson, D. Haas, S. Riddler, R. Haubrich, D. Heckerman, S. Mallal, and M. John.  Translation of HLA-HIV associations to the cellular level: HIV adapts to inflate CD8 T cell responses against Nef and HLA-adapted variant epitopesJournal of Immunology, 187(5):2502-13, Aug, 2011.

 

·        E. Lazaro, C. Kadie, P. Stamegna, S. Zhang, P. Gourdain, N. Lai, M. Zhang, S. Martinez, D. Heckerman, and S. Le Gall.  Variable HIV peptide stability in human cytosol is critical to epitope presentation and immune escapeJ Clin Invest., 121(6): 2480-2492, June 2011.

 

·        N. Keane, S. Roberts, C. Almeida, T. Krishnan, A. Chopra, E. Demaine, R. Laird, M. Tschochner, J. Carlson, S. Mallal, D. Heckerman, I. James, and M. John.  High-avidity, high-IFNg-producing CD8 T-cell responses following immune selection during HIV-1 infectionImmunology and Cell Biology, doi:10.1038/icb.2011.34, May, 2011.

 

·        P. Matthews, E. Adland, J. Listgarten, A. Leslie, N. Mkhwanazi, J. Carlson, M. Harndahl, A. Stryhn, R. Payne, A. Ogwu, K. Huang, J. Frater, P. Paioni, H. Kloverpris, P.Jooste, D. Goedhals, C. van Vuuren, D. Steyn, L. Riddell, F. Chen, G. Luzzi, T. Balachandran, T. Ndung'u, S. Buus, M. Carrington, R. Shapiro, D. Heckerman, and P. Goulder.  HLA-A*7401-mediated control of HIV viremia is independent of its linkage disequilibrium with HLA-B*5703Journal of Immunology, doi: 10.4049, April 2011.

 

·        K. Huang, D. Goedhals, J. Carlson, M. Brockman, S. Mishra, Z. Brumme, S. Hickling, C. Tang, T. Miura7, C. Seebregts, D. Heckerman, T. Ndung'u, B. Walker, P. Klenerman, D. Steyn, P. Goulder, R. Phillips, Bloemfontein-Oxford Collaborative Group, C. van Vuuren, and J. Frater.  Progression to AIDS in South Africa is associated with both reverting and compensatory viral mutationsPLoS One, 6(4):e19018, April, 2011.

 

·        M. Rolland, N. Frahm, D. Nickle, N. Jojic, W. Deng, T. Allen, C. Brander, D. Heckerman, J. Mullins.  Increased breadth and depth of cytotoxic T lymphocytes responses against HIV-1-B Nef by inclusion of epitope variant sequencesPLoS One, 28;6(3):e17969, March 2011.

 

·        J. Wright, V. Novitsky, M. Brockman, Z. Brumme, C. Brumme, J. Carlson, D. Heckerman, B. Wang, E. Losina, M. Leshwedi, M. van der Stok, L. Maphumulo, N. Mkhwanazi, F. Chonco, P. Goulder, M. Essex, B. Walker, and T. Ndung'u.  Influence of Gag-Protease-mediated replication vapacity on disease progression in individuals recently infected with HIV-1 subtype C.  Journal of Virology, doi:10.1128/JVI.02520-10, February, 2010.

 

·        The International HIV Controllers Study.  The Major Genetic Determinants of HIV-1 Control Affect HLA Class I Peptide Presentation, Science, 330: 1551-1557, December 10 2010.

 

·        H. Crawford, P. Matthews, M. Schaefer, J. Carlson, A. Leslie, W. Kilembe, S. Allen, T. Ndung'u, D. Heckerman, E. Hunter, and P. Goulder.  The hypervariable HIV-1 capsid protein residues comprise HLA-driven CD8+ T-cell escape mutations and covarying HLA-independent polymorphismsJournal of Virology, doi:10.1128/JVI.01879-10, November, 2010.

 

·        M. Brockman, Z. Brumme, C. Brumme, T. Miura, J. Sela, P. Rosato, C. Kadie, J. Carlson, T. Markle, H. Streeck, A. Kelleher, M. Markowitz, H. Jessen, E. Rosenberg, M. Altfeld, P. Harrigan, D. Heckerman, B. Walker, and T. Allen.  Early selection in Gag by protective HLA alleles contributes to reduced HIV-1 replication capacity that may be largely compensated for in chronic infectionJournal of Virology, 84(22): 11937-11949, November 2010.

 

·        A. Leslie, P. Matthews, J. Listgarten, J. Carlson, C. Kadie, T. Ndung'u, C. Brander, H. Coovadia, B. Walker, D. Heckerman and P. Goulder. Additive contribution of HLA class I alleles in the immune control of HIV-1 infection. Journal of Virology, 84(19):9879-9888, October 2010.

 

·        H. Laaksovirta, T. Peuralinna, J. Schymick, S. Scholz, S. Lai, L. Myllykangas, R. Sulkava, L. Jansson, D. Hernandez, J. Gibbs, M. Nalls, D. Heckerman, P. Tienari, B. Traynor.  Chromosome 9p21 in amyotrophic lateral sclerosis in Finland: A genome-wide association studyThe Lancet Neurology, 9(10): 978-985, October 2010.

 

·        J. Listgarten, C. Kadie, E. Schadt, D. Heckerman.  Correction for hidden confounders in the genetic analysis of gene expression. PNAS, 107 (38): 16465-16470, September 2010 (doi: 10.1073/pnas.1002425107).

 

·        M. Rolland, J. Carlson, S. Manocheewa, J. Swain, E. Lanxon-Cookson, W. Deng,  C. Rousseau, D. Raugi, G. Learn, B. Maust, H. Coovadia, T. Ndung'u, P. Goulder, B. Walker, C. Brander, D. Heckerman, J. Mullins.  Amino-acid co-variation in HIV-1 Gag subtype C: HLA-mediated selection pressure and compensatory dynamicsPLoS One, 5(9), September 1, 2010.

 

·        A. S, V.Y.F. Tan, J. Winn, M. Svens, C.M. Bishop, D. Heckerman, I. Buchan, and A. Custovic, Beyond atopy: Multiple patterns of sensitization in relation to asthma in a birth cohort studyAmerican Journal of Respiratory and Critical Care Medicine, 181(11):1200-6, June 2010.

 

·        R. Shachter and D. Heckerman.  Pearl Causality and the Value of Control.  In Heuristics, Probability and Causality: A Tribute to Judea Pearl, R. Dechter, H. Geffner, and J. Halpern eds.  2010.  Local copy.

 

·        M. John, D. Heckerman, I. James, L. Park, J. Carlson, A. Chopra, S. Gaudieri, D. Nolan, D. Haas, S. Riddler, R. Haubrich, and Simon Mallal.  Adaptive interactions between HLA and HIV-1: Highly divergent selection imposed by HLA class I molecules with common supertype motifsThe Journal of Immunology, 184:4368-4377, March 2010.

 

·        M. Koga, A. Kawana-Tachikawa, D. Heckerman, T. Odawara, H. Nakamura, T. Koibuchi, T. Fujii, T. Miura, and A. Iwamoto.  Changes in impact of HLA class I allele expression on HIV-1 plasma virus loads at a population level over timeMicrobiology and Immunology, 54(4): 196-205.  January 2010.

 

·        A. Bansal, J. Carlson, J. Yan, O. Akinsiku, M. Schaefer, S. Sabbaj, A. Bet, D. Levy, S. Heath, J. Tang, R. Kaslow, B. Walker, T. Ndungu, P. Goulder, D. Heckerman, E. Hunter, and P. Goepfert.  CD8 T cell response and evolutionary pressure to HIV-1 cryptic epitopes derived from antisense transcriptionJEM, 10.1084/jem.20092060, January 2010.

 

·        C. Berger, J. Carlson, C. Brumme, K. Hartman, Z. Brumme, L. Henry, P. Rosato, A. Piechocka-Trocha, M. Brockman, P. Harrigan, D. Heckerman, D. Kaufmann, and Ch. Brander.  Viral adaptation to immune selection pressure by HLA class I-restricted CTL responses targeting epitopes in HIV frameshift sequencesJEM, 10.1084/jem.20091808, January 2010.

 

·        B. Lee, L. Nachmanson, G. Robertson, J. Carlson, and D. Heckerman.  PhyloDet: A scalable visualization tool for mapping multiple traits to large evolutionary treesBioinformatics, 25(19):2611-2612.  October 2009.

 

·        S. Avila-Rios, C. Ormsby, J. Carlson, H. Valenzuela-Ponce, J. Blanco-Heredia, D. Garrido-Rodriguez, C. Garcia-Morales, D. Heckerman, Z. Brumme, S. Mallal, M. John, E. Espinosa, and G. Reyes-Teran.  Unique features of HLA-mediated HIV evolution in a Mexican cohort: a comparative studyRetrovirology, 6:72doi:10.1186/1742-4690-6-72, August 2009.

 

·        Z. Brumme, M. John., J. Carlson, C. Brumme, D. Chan, M. Brockman, L. Swenson, I. Tao, S. Szeto, P. Rosato, J. Sela, C. Kadie, N. Frahm, C. Brander, D. Haas, S. Riddler, R. Haubrich, B. Walker, P. Harrigan, D. Heckerman, and S. Mallal.  HLA-associated immune escape pathways in HIV-1 subtype B Gag, Pol and Nef proteins. PLoS ONE, 4(8):e6687. doi:10.1371/journal.pone.0006687.  August 2009.

 

·        C. Rousseau, D. Lockhart, J. Listgarten, S. Maley, C. Kadie, G. Learn, D. Nickle, D. Heckerman, W. Deng, C. Brander, T. Ndung'u, H. Coovadia, P. Goulder, B. Korber, B. Walker, J. Mullins.  Rare HLA drive additional HIV evolution compared to more frequent alleles.  AIDS Res Hum Retroviruses, 25:297-303, March 2009.

 

·        Y. Kawashima, K. Pfafferott, J. Frater, P. Matthews, R. Payne, M. Addo, H. Gatanaga, M. Fujiwara, A. Hachiya, H. Koizumi, N. Kuse, S. Oka, A. Duda, A. Prendergast, H. Crawford, A. Leslie, Z. Brumme, C. Brumme, T. Allen, C. Brander, R. Kaslow, J. Tang, E. Hunter, S. Allen, J. Mulenga, S. Branch, T. Roach, M. John, S. Mallal, A. Ogwu, R. Shapiro, J. Prado, S. Fidler, J. Weber, O. Pybus, P. Klenerman, T. Ndung'u, R. Phillips, D. Heckerman, P. Harrigan, B. Walker, M. Takiguchi, and P. Goulder.  Adaptation of HIV-1 to human leukocyte antigen class I.  Nature, February 2009.

 

·        T. Miura, M. Brockman, Z. Brumme, C. Brumme, F. Pereyra, A. Trocha, B. Block, A. Schneidewind, T. Allen, D. Heckerman, and B. Walker.  HLA-associated alterations in replication capacity of chimeric NL4-3 viruses carrying gag-protease from elite controllers of human immunodeficiency virus type 1.  Journal of Virology, 83:140-149. January 2009.

 

·        J. Carlson, Z. Brumme, C. Rousseau, C. Brumme, P. Matthews, C. Kadie, J. Mullins, B. Walker, P. Harrigan, P. Goulder, D. Heckerman.  Phylogenetic dependency networks: Inferring patterns of CTL escape and codon covariation in HIV-1 Gag. PLoS Computational Biology, 4(11): e1000225, November 2008.

 

      Further develops the mixed model of Bhattacharya et al. (Science 2007) to correct for HIV-sequence relatedness when examining associations between the HLA type of an infected individual and the amino-acid polymorphisms in the infecting HIV sequence.

 

·       Y. Wang, B. Li, J. Carlson, H. Streeck, A. Gladden, R. Goodman, A. Schneidewind, K. Power, I. Toth, N. Frahm, G. Alter, C. Brander, M. Carrington, B. Walker, M. Altfeld, D. Heckerman, and T. Allen.  Protective HLA class I alleles restricting acute-phase CD8+ T cell responses are associated with viral escape mutations located in highly conserved regions of HIV-1Journal of Virology.  November, 2008.

 

·        D. Yerly, D. Heckerman, T. Allen, T. Suscovich, N. Jojic, C. Kadie, W. Pichler, A. Cerny, and C. Brander.  Design, expression, and processing of epitomized hepatitis C virus-encoded CTL epitopes.  Journal of Immunology, 181:6361-6370, November, 2008.

 

·        N. Zaitlen, M. Reyes-Gomez, D. Heckerman, N. Jojic. Shift-invariant adaptive double threading: Learning MHC II-peptide binding. J Comput Biol., 15:927-42, September, 2008.

 

·        P. Matthews, A. Prendergast, A. Leslie, H. Crawford, R. Payne, C. Rousseau, I. Honeyborne, J. Carlson, C. Kadie, C. Brander, J. Mullins, H. Coovadia, T. Nding.u, B. Walker, D. Heckerman, P. Goulder. Central role of reverting mutations in HLA associations with human immunodeficiency virus set point. Journal of Virology, 82:8548-59, September, 2008.

 

·        Z. Brumme, C. Brumme, J. Carlson, H. Streeck, M. John, Q. Eichbaum, B. Block, B. Baker, C. Kadie, M. Markowitz, H. Jessen, A. Kelleher, E. Rosenberg, J. Kaldor, Y. Yuki, M. Carrington, T. Allen, S. Mallal, M. Altfeld, D. Heckerman, and B. Walker.  Marked epitope- and allele-specific differences in rates of mutation in human immunodeficiency type 1 (HIV-1) Gag, Pol, and Nef cytotoxic T-lymphocyte epitopes in acute/early HIV-1 infectionJournal of Virology, 82:9216-9227, September 2008.

 

·        T. Miura, M. Brockman, C. Brumme, Z. Brumme, J. Carlson, F. Pereyra, A. Trocha, M. Addo, B. Block, A. Rothchild, B. Baker, T. Flynn, A. Schneidewind, B. Li, Y. Wang, D. Heckerman, T. Allen, and B. Walker.  Genetic characterization of human immunodeficiency virus type 1 in elite controllers: Lack of gross genetic defects or common amino acid changes.  Journal of Virology, 82:8422-8430, September, 2008.

 

·        T. Kuntzen, J. Timm, A. Berical, N. Lennon, A. Berlin, S. Young, B. Lee, D. Heckerman, J. Carlson, L. Reyor, M. Kleyman, C. McMahon, C. Birch, J. Schulze Zur Wiesch, T. Ledlie, M. Koehrsen, C. Kodira, A. Roberts, G. Lauer, H. Rosen, F. Bihl, A. Cerny, U. Spengler, Z. Liu, A. Kim, Y. Xing, A. Schneidewind, M. Madey, J. Fleckenstein, V. Park, J. Galagan, C. Nusbaum, B. Walker, G. Lake-Bakaar, E. Daar, I. Jacobson, E. Gomperts, B. Edlin, S. Donfield, R. Chung, A. Talal, T. Marion, B. Birren, M. Henn, T. Allen.  Naturally occurring dominant resistance mutations to hepatitis C virus protease and polymerase inhibitors in treatment-naïve patients.  Hepatology, 48:1769-1778, July 2008.

 

·       C. Wang, D. Blei, and D. Heckerman.  Continuous Time Dynamic Topic Models.  In Proceedings of Twenty Fourth Conference on Uncertainty in Artificial Intelligence, Helsinki, Finland, UAI Press, July 2008.  Local copy.

 

·        Z. Brumme., I. Tao, S. Szeto, C. Brumme, J. Carlson, D. Chan, C. Kadie, N. Frahm, C. Brander, B. Walker, D. Heckerman, and P. Harrigan.  HLA-specific polymorphisms in HIV-1 Gag and their association with viral load in chronic untreated, AIDS, 22(11):1277-1286, July, 2008.

 

·         P. Goepfert, W. Lumm, P. Farmer, P. Matthews, A. Prendergast, J. Carlson, C. Derdeyn, J. Tang, R. Kaslow, A. Bansal, K. Yusim, D. Heckerman, J. Mulenga, S. Allen, P. Goulder, and E. Hunter.  Transmission of HIV-1 Gag immune escape mutations is associated with reduced viral load in linked recipients, Journal of Experimental Medicine, 205(5):1009-1017, April, 2008.

 

·        C. Rousseau, M. Daniels, J. Carlson C. Kadie, H. Crawford, A. Prendergast, P. Matthews, D. Raugi, B. Maust G. Learn D. Nickle N. Frahm, C. Brander, B. Walker P. Goulder, T. Bhattacharya, D. Heckerman, B. Korber, and J. Mullins.  Class-I driven evolution of human immunodeficiency virus type 1 subtype C proteome: Immune escape and viral load, Journal of Virology, doi:10.1128/JVI.02455-07, April 2008.

 

·        H. Kang, N. Zaitlen, C. Wade, A. Kirby, D. Heckerman, M. Daly, and E. Eskin, Efficient Control of Population Structure in Model Organism Association Mapping, Genetics, 178:1709-1723, March, 2008 (doi: 10.1534/genetics.107.080101).

 

Applies linear mixed models to GWAS on human data.  The work includes computational speedups over the naïve algorithm as well as a phylogeny-based similarity matrix.

 

Historical note: This work arose from that on PhyloD (see below).  With respect to the model representing similarity among samples, I replaced the phylogenetic tree (more appropriate for monoploid organisms) with a multivariate Gaussian (more appropriate for polyploid organisms), unknowingly reinventing linear mixed models.  I discussed the model with Eleazar, who in turn discussed it with Nicholas Schork, who noted the correspondence.  Eleazar and team then went on to develop the computational efficiencies described in the paper.

 

·        J. Listgarten, Z. Brumme, C. Kadie, G. Xiaojiang, B. Walker, M. Carrington, P. Goulder, and D. Heckerman.  Statistical resolution of ambiguous HLA typing dataPLoS Computational Biology, 4(2): e1000016, February, 2008.

 

·        D. Yerly, D. Heckerman, T. Allen, J. Chisholm, K. Faircloth, C. Linde, N. Frahm, J. Timm, W. Pichler, A. Cerny, and C. Brander.  Increased CTL epitope variant cross-recognition and functional avidity are associated with HCV clearance.  J. Virol, January 2008.

 

·        D. Nickle, N. Jojic, D. Heckerman, V. Jojic, D. Kirovski, M. Rolland, S. Pond, J. Mullins.  Comparison of immunogen designs that optimize peptide coverage: Reply to Fischer et al., PLoS Computational Biology, 4(1):e25, January 2008.

 

·        M. Rolland, D. Heckerman, W. Deng, C. Rousseau, H. Coovadia K. Bishop, P. Goulder, B. Walker, C. Brander, J. Mullins.  Broad and Gag-biased HIV-1 epitope repertoires are associated with lower viral loads. PLoS ONE, 3(1): e1424, January, 2008.

 

·        K.C. Ngumbela, C. Day, Z. Mncube, K. Nair, D. Ramduth, C. Thobakgale, E. Moodley, S. Reddy, C. de Pierres, N. Mkhwanazi, K. Bishop, M. van der Stok, N. Ismail, I. Honeyborne, H. Crawford, D. Kavanagh, C. Rousseau, D. Nickle, J. Mullins, D. Heckerman, B. Korber, H. Coovadia, P. Goulder, and B. Walker.  Targeting of a CD8 T cell Env epitope presented by HLA-B*5802 is associated with markers of HIV disease progression and lack of selection pressure, AIDS Res Hum Retroviruses. 2008 Jan;24(1):72-82.

 

·        J. Listgarten, N. Frahm, C. Kadie, C. Brander, D. Heckerman.  A statistical framework for modeling HLA-dependent T cell response data, PLoS Computational Biology, 3(10): e188, October 2007.

 

·        D. Heckerman, C. Kadie, and J. Listgarten.  Leveraging information across HLA alleles/supertypes improves epitope prediction.  J. of Comp. Bio, 14(6): 736-746, August 2007.  Microsoft copy.

 

·        N. Frahm, K. Yusim, T. Suscovich, S. Adams, J. Sidney, P. Hraber, H. Hewitt, Ca. Linde, D. Kavanagh, T. Woodberry, L. Henry, K. Faircloth, J. Listgarten, C. Kadie, N. Jojic, K. Sango, N. Brown, E. Pae, M. Zaman, F. Bihl, A. Khatri, M. John, S. Mallal, F. Marincola, B. Walker, A. Sette, D. Heckerman, B. Korber, C. Brander.  Extensive HLA class I allele promiscuity among viral CTL epitopes. EJI, 37, August 2007.

 

·        Z. Brumme, C. Brumme, D. Heckerman, B. Korber, M. Daniels, J. Carlson, C. Kadie, T. Bhattacharya, C. Chui, T. Mo, R. Hogg, J. Montaner, N. Frahm, C. Brander, B. Walker, P. Harrigan.  Evidence of Differential HLA Class I-Mediated Viral Evolution in Functional and Accessory/Regulatory Genes of HIV-1. PLoS Pathogens, 3(7): e94, July 2007.

 

·        J. Carlson, C. Kadie, S. Mallal, and D. Heckerman.  Leveraging hierarchical population structure in discrete association studies. PLoS ONE, 2(7): e591, July 2007.

 

      Further develops the mixed model of Bhattacharya et al. (Science 2007) to correct for HIV-sequence relatedness when examining associations between the HLA type of an infected individual and the amino-acid polymorphisms in the infecting HIV sequence.

 

·        J. Listgarten and D. Heckerman, Determining the number of non-spurious arcs in a learned DAG model: Investigation of a Bayesian and a frequentist approach.  In Proceedings of Twenty Third Conference on Uncertainty in Artificial Intelligence, Vancouver, Canada, UAI Press, July 2007.  Microsoft copyLocal copy.

 

·        C. Rousseau, G. Learn, T. Bhattacharya, D. Nickle, D. Heckerman, S. Chetty, C. Brander, P. Goulder, B. Walker, P. Kiepiela, B. Korber, and J. Mullins.  Extensive intrasubtype recombination in South African Human Immunodeficiency Virus type I subtype C infections.  J Virol, 81(9): 4492-4500.  May, 2007.

 

·        D. Nickle, M. Rolland, M. Jensen, S. Pond, W. Deng, M. Seligman, D. Heckerman, J. Mullins, and N. Jojic.  Coping with viral diversity in HIV vaccine design, PLoS Computational Biology, 3(4): e75, April 2007.

 

·        T. Bhattacharya, M. Daniels, D. Heckerman, B. Foley, N. Frahm, C. Kadie, J. Carlson, K. Yusim, B. McMahon, B. Gaschen, S. Mallal, J. Mullins, D. Nickle, J. Herbeck, C. Rousseau, G. Learn, T. Miura, C. Brander, B. Walker, B. Korber.  Founder effects in the assessment of HIV polymorphisms and HLA allele associations, Science, 315, 1583-1586, March 16 2007.  Talk at ICML 2007.

 

Describes PhyloD (“second method” in the supplement), a mixed model with a discrete-valued random effect and shows how it corrects for HIV-sequence relatedness when examining associations between the HLA type of an infected individual and the amino-acid polymorphisms in the infecting HIV sequence.  As described above, this work led to our work on linear mixed models.

 

·        J. Goodman, G.V. Cormack, D. Heckerman, Spam and the ongoing battle for the inbox. Communications of the ACM, Vol. 50 No. 2, Pages 24-33, February 2007, 10.1145/1216016.1216017.

 

·        P. Kiepiela, K. Ngumbela, C. Thobakgale, D. Ramduth, I. Honeyborne, E. Moodley, S. Reddy, C. de Pierres, Z. Mncube, N. Mkhwanazi, K. Bishop, M. van der Stok, K. Nair, N. Khan, H. Crawford, R. Payne, A. Leslie, J. Prado, A. Prendergast, J. Frater, N. McCarthy, C. Brander, G. Learn, D. Nickle, C. Rousseau, H. Coovadia, J. Mullins, D. Heckerman, B. Walker, and P. Goulder.  CD8+ T-cell responses to different HIV proteins have discordant associations with viral load, Nature Medicine, December 17 2006.

 

·        F. Bach, D. Heckerman, E. Horvitz, Considering cost asymmetry in learning classifiersJournal of Machine Learning Research, 7, 1713-1741, 2006.  Microsoft copy.

 

·        N. Jojic, V. Jojic, B. Frey, C. Meek, and D. Heckerman.  Using epitomes to model genetic diversity: Rational design of HIV vaccine cocktails.  NIPS 2005.  Local copy.

 

·        J. Goodman, D. Heckerman, and R. Rounthwaite.  Stopping SpamScientific American, April, 2005.  Microsoft copy.

 

·        G. Shani, D. Heckerman, and R. Brafman.  An MDP-based recommender systemJournal of Machine Learning Research 6: 1265-1295, 2005.  UAI 2002 version (Microsoft copy; local copy).

 

·        F. Bach, D. Heckerman, and E. Horvitz, On the Path to an Ideal ROC Curve: Considering Cost Asymmetry in Learning Classifiers.  In Proceedings of the Tenth International Workshop on Artificial Intelligence and Statistics, Barbados, January 2005.  Microsoft copy.  Local copy.

 

·        D. Chickering, D. Heckerman, and C. Meek, Large-Sample Learning of Bayesian Networks is NP-HardJournal of Machine Learning Research. 5: 1287-1330, 2004.  Microsoft copy.

 

In comparison to Geiger and Heckerman, arXiv 1993, this paper starts with stronger preconditions (local distributions are binary-like lattice and binary-like totally strictly positive) and proves a stronger result (joint distribution is perfect with respect to the graph).

 

·        V. Jojic, N. Jojic, C. Meek, D. Geiger, A. Siepel, D. Haussler, and D. Heckerman: Efficient approximations for learning phylogenetic HMM models from data. ISMB/ECCB (Supplement of Bioinformatics) 161-168, 2004.  Microsoft copy.

 

·        N. Jojic, V. Jojic, and D. Heckerman, Joint discovery of haplotype blocks and complex trait associations from SNP sequences.  In Proceedings of Twentieth Conference on Uncertainty in Artificial Intelligence, Banff, Canada, UAI Press, July 2004.  Microsoft copy.  Local copy.

 

·        B. Thiesson, D. Chickering, D. Heckerman, and C. Meek, ARMA Time-Series Modeling with Graphical Models.  In Proceedings of Twentieth Conference on Uncertainty in Artificial Intelligence, Banff, Canada, UAI Press, July 2004.  Microsoft copyLocal copy.

 

·        D. Heckerman, C. Meek, and D. Koller, Probabilistic Entity-Relationship Models, PRMs and Plate Models.  In L. Getoor and B. Taskar, editors, Introduction to Statistical Relational Learning, MIT Press, 2007.  CiteSeer copy.  Microsoft copy.  Local copy (2004 version).

 

·        D. Chickering and D Heckerman. Targeted Advertising with Inventory Management.  Interfaces, 33:71-77, 2003.  Microsoft copy.  Local copy.

 

·        I. Cadez, D. Heckerman, C. Meek, P. Smyth, and S. White, Visualization of Navigation Patterns on a Web Site Using Model Based Clustering, Data Mining and Knowledge Discovery, 7:399-424, 2003.  Microsoft copy.  Local copy.

 

·        G. Hulten, D.M. Chickering, D. Heckerman. Learning Bayesian Networks from Dependency Networks: A Preliminary Study. In Proceedings of the Ninth International Workshop on Artificial Intelligence and Statistics, Key West, FL, January 2003.  Microsoft copy.

 

·        C. Meek, B. Thiesson, and D. Heckerman,  Staged Mixture Modeling and Boosting.  In Proceedings of Eighteenth Conference on Uncertainty in Artificial Intelligence, Edmonton, Alberta, Morgan Kaufmann, August 2002.  Microsoft copyLocal copy.

 

·       C. Kadie, C. Meek, and D. Heckerman.  CFW: A collaborative filtering system using posteriors over weights of evidence.  In Proceedings of Eighteenth Conference on Uncertainty in Artificial Intelligence, Edmonton, Alberta, Morgan Kaufmann, August 2002.  Microsoft copyLocal copy.

 

·        C. Meek, D. Chickering and D. Heckerman.  Autoregressive tree models for time-series analysis. In Proceedings of the Second International SIAM Conference on Data Mining, Arlington, VA, SIAM, April, 2002.  Microsoft copy.

 

·        D. Geiger and D. Heckerman. Parameter Priors for Directed Acyclic Graphical Models and the Characterization of Several Probability Distributions.  The Annals of Statistics, 30: 1412-1440, 2002.  UAI 1999 version.  Annals version.

 

·        C. Meek, B. Thiesson, and D. Heckerman.  The Learning-Curve Sampling Method Applied to Model-Based Clustering.  Journal of Machine Learning Research, 2:397-418, 2001.  Microsoft copy.

 

·         N. Jojic, P. Simard, B. J. Frey and D. Heckerman. Learning mixtures of smooth, nonuniform deformation models for probabilistic image matching.  In Proceedings of Eighth International Workshop on Artificial Intelligence and Statistics, Key West, FL, Morgan Kaufmann, January 2001.  Microsoft copy.  Local copy.

 

·        B. Thiesson, C. Meek, and D. Heckerman. Accelerating EM for Large Databases Machine Learning, 45:279-299, 2001.  Microsoft copy.

 

·        D. Chickering and D. Heckerman. A Decision-Theoretic Approach to Targeted Advertising.  In Proceedings of Sixteenth Conference on Uncertainty in Artificial Intelligence, Stanford, CA, Morgan Kaufmann, July 2000.  Microsoft copy.  Local copy.

 

·        D. Heckerman, D. Chickering, C. Meek, R. Rounthwaite, C. Kadie. Dependency Networks for Density Estimation, Collaborative Filtering, and Data VisualizationJournal of Machine Learning Research. 1:49-75, 2000.  Microsoft copy.

 

·        D. Heckerman and D. Chickering. A Comparison of Scientific and Engineering Criteria for Bayesian Model SelectionStatistics and Computing, 10:55-62, 2000.  Microsoft copy.  Local copy.  Earlier version in In Proceedings of the Sixth International Workshop on Artificial Intelligence and StatisticsPMLR R1:275-282, 1997.

 

·        D. Chickering and D. Heckerman. Fast Learning from Sparse Data.  In Proceedings of Fifteenth Conference on Uncertainty in Artificial Intelligence, Stockholm, Sweden, Morgan Kaufmann, August 1999.  Microsoft copyLocal copy.

 

·        D. Heckerman, C. Meek, and G. Cooper A Bayesian Approach to Causal Discovery.  In C. Glymour and G. Cooper, editors, Computation, Causation, and Discovery, pages 141-165.  MIT Press, Cambridge, MA, 1999.  Local copy.  Also in Spirtes et al., Causation, Prediction, and Search, Second Edition, MIT Press 2000.

 

·        D. Geiger, D. Heckerman, H. King, C. Meek. Stratified Exponential Families: Graphical Models and Model Selection.  The Annals of Statistics, 29:505-529, 2001.  A preliminary version can be found in the Proceedings of Fifth Conference on Artificial Intelligence and Statistics, Ft. Lauderdale, FL. January 1999.  Microsoft copy.

 

·        S. T. Dumais, J. Platt, D. Heckerman and M. Sahami. Inductive Learning Algorithms and Representations for Text Categorization.  Proceedings of ACM-CIKM98, November, 1998.  Microsoft copy.  Local copy.  Winner of CIKM Test of Time Award 2017.

 

      Historical note: This work made its way into versions of Microsoft Sharepoint Portal Server.

 

·        M. Sahami, S. Dumais, D. Heckerman, and E. Horvitz. A Bayesian Approach to Filtering Junk E-mail. AAAI'98 Workshop on Learning for Text Categorization, July 27, 1998, Madison, Wisconsin.  Microsoft copy.  Local copy.  A talk on the history of junk-mail filtering and its relationship to designing a vaccine for HIV was given at the Microsoft Worldwide Partner Conference 2013.

 

      Historical note: On January 17, 1997, after receiving an unprecedented three junk mails within two days, I wrote to my team mates Eric Horvitz and Jack Breese, “I've had it with junk mail...let's build a Bayes net (with learning of course) that identifies mail as being junk!”  Later that year, Mehran joined Microsoft Research as an intern, and we built the first filter.

       

·        D. Heckerman and E. Horvitz. Inferring Informational Goals from Free-Text Queries. In Proceedings of Fourteenth Conference on Uncertainty in Artificial Intelligence, Madison, WI, Morgan Kaufmann, July 1998.  Microsoft copy.  Local copy.

 

      Historical note: This is a look back and the invention of the Answer Wizard, which became the backend for the Office Assistant (aka “Clippy”) included in Microsoft Office for Windows (versions 97 to 2003).  I invented the first version of the Answer Wizard just months after arriving at Microsoft Research in 1992, when I couldn’t find help in Excel on the use of graphs.  After a lot of asking around, I finally learned that graphs were called “charts.”

 

·        E. Horvitz, J. Breese, D. Heckerman, D. Hovel, and K. Rommelse. The Lumiere Project: Bayesian User Modeling for Inferring the Goals and Needs of Software Users. In Proceedings of Fourteenth Conference on Uncertainty in Artificial Intelligence, Madison, WI, Morgan Kaufmann, July 1998.  Microsoft copy.  Local copy.

 

·        J. Breese, D. Heckerman, C. Kadie Empirical Analysis of Predictive Algorithms for Collaborative Filtering.  In Proceedings of Fourteenth Conference on Uncertainty in Artificial Intelligence, Madison, WI, Morgan Kaufmann, July 1998.  Microsoft copyLocal copy.

 

      Historical note: This work appears to be the first time that the task of collaborative filtering was cast as a standard prediction problem, amenable to a multitude of machine learning algorithms.  Based on this work, the Machine Learning and Applied Statistics Group at Microsoft Research built a collaborative-filtering engine for Microsoft Commerce Server, which first shipped in 2000.

 

·        M. Meila and D. Heckerman An Experimental Comparison of Several Clustering and Initialization MethodsMachine Learning.  42:9-29, 2001.  Microsoft copy.  UAI 1998 version (local copy).

 

·        B. Thiesson, C. Meek, D. Chickering, D. Heckerman.  Computationally Efficient Methods for Selecting Among Mixtures of Graphical Models.  In J. M. Bernardo, J. O. Berger, A. P. Dawid, and A. F. M. Smith, editors, Bayesian Statistics 6, pages 631-656, Oxford University Press, Oxford, 1999.  Local copyUAI 1998 version (Microsoft copy; local copy).

 

·        D. Heckerman and C. Meek. Models and Selection Criteria for Regression and Classification.  In Proceedings of Thirteenth Conference on Uncertainty in Artificial Intelligence, Providence, RI, pages 223-228. Morgan Kaufmann, August 1997.  Microsoft copy.  Local copy.

 

·        D. Chickering, D. Heckerman, C. Meek.  A Bayesian Approach to Learning Bayesian Networks with Local Structure.  In Proceedings of Thirteenth Conference on Uncertainty in Artificial Intelligence, Providence, RI, pages 80-89. Morgan Kaufmann, August 1997.  Microsoft copyLocal copy.

 

·        C. Meek and D. Heckerman. Structure and parameter learning for causal independence and causal interaction models.  In Proceedings of Thirteenth Conference on Uncertainty in Artificial Intelligence, Providence, RI, pages 366-375. Morgan Kaufmann, August 1997.  Microsoft copy.  Local copy.

 

·        D. Heckerman and C. Meek Embedded Bayesian Network Classifiers.  Technical Report MSR-TR-97-06, Microsoft Research, March, 1997.  Microsoft copy.  Local copy.

 

·        P. Smyth, D. Heckerman, M. Jordan. Probabilistic Independence Networks for Hidden Markov Probability ModelsNeural Computation, 9:227-269, 1997.  Microsoft copy.  Local copy.

 

·        D. Chickering and D. Heckerman. Efficient Approximations for the Marginal Likelihood of Bayesian Networks With Hidden VariablesMachine Learning, 29:181-212, 1997.  UAI 1996 version (local copy). 

 

·        D. Geiger, D. Heckerman, and C. Meek. Asymptotic Model Selection for Directed Networks with Hidden Variables.  In Proceedings of Twelfth Conference on Uncertainty in Artificial Intelligence, Portland, OR, pages 283-290. Morgan Kaufmann, August 1996.  Microsoft copyLocal copy.

 

·        J. Breese and D. Heckerman. Decision-theoretic troubleshooting: A framework for repair and experiment.  In Proceedings of Twelfth Conference on Uncertainty in Artificial Intelligence, Portland, OR, pages 124-132. Morgan Kaufmann, August 1996.  Microsoft copy.  Local copy.

 

·        J. Breese and D. Heckerman.  Decision-theoretic case-based reasoning.  IEEE Transactions on Systems, Man, and Cybernetics, 26:838-842, 1996.  Microsoft copy.  Local copy.

 

·        D. Geiger, D. Heckerman. A Characterization of the Dirichlet Distribution Through Global and Local Independence.  The Annals of Statistics, 25:1344-1369, 1997.  Microsoft copy.

 

·        D. Geiger and D. Heckerman.  Knowledge representation and inference in similarity networks and Bayesian multinetsArtificial Intelligence, 82:45-74, 1996.  Local copy.

 

·        D. Heckerman, J. Breese. Causal Independence for Probability Assessment and Inference Using Bayesian Networks.  IEEE Transactions on Systems, Man, and Cybernetics, 26:826-831, 1996.  Microsoft copy.  Local copy.

 

·        D. Geiger, D. Heckerman. A Characterization of the Bivariate Normal-Wishart Distribution. Probability and Mathematical Statistics, 18:119-131, 1998.  Microsoft copy.

 

·        D. Heckerman, D. Geiger. Likelihoods and Parameter Priors for Bayesian Networks. MSR-TR-95-54, Nov, 1995.  arXiv:2105.06241, May, 2021.

 

·        D. Heckerman, R. Shachter. Decision-Theoretic Foundations for Causal Reasoning.  Journal of Artificial Intelligence Research, 3:405-430, 1995.  Microsoft Copy.  UAI 1995 version (local copy).

      Provides definitions of cause and effect based on Savage’s theory of decision making.  In addition, connects the Savage view with the view expressed in the works of Wold, Spirtes, Glymour, and Scheines, and Pearl.  A notable realization is that the notion of setting versus observing a variable are just as difficult to define as the notions of cause and effect.

 

·        D. Heckerman, D. Geiger, D. Chickering. Learning Bayesian networks: The Combination of Knowledge and Statistical Data.  Machine Learning, 20:197-243, 1995.  Microsoft copy.

 

·        D. Chickering, D. Geiger, and D. Heckerman.  On finding a cycle basis with a shortest maximal cycleInformation Processing Letters, 54:55-58, 1995.  Microsoft copy.  Local copy.

 

·        D. Heckerman. A Bayesian approach to learning causal networks.  In Proceedings of Eleventh Conference on Uncertainty in Artificial Intelligence, Montreal, Quebec, pages 285-295. Morgan Kaufmann, August 1995.  Microsoft copyLocal copy.

 

      This paper identifies sufficient assumptions for learning causal models from data with a combination of observational and interventional data.  Section 7 deals with learning causal models with missing data.  One consequence of the results in this section is that parameter modularity is not conjugate with respect to missing data.

 

·        D. Geiger and D. Heckerman. A characterization of the Dirichlet distribution with applications to learning Bayesian networks.  In Proceedings of Eleventh Conference on Uncertainty in Artificial Intelligence, Montreal, Quebec, pages 196-207. Morgan Kaufmann, August 1995.

 

·        D. Chickering, D. Geiger, and D. Heckerman. Learning Bayesian Networks: Search Methods and Experimental Results.  In Proceedings of Fifth Conference on Artificial Intelligence and Statistics, Ft. Lauderdale, FL. Society for Artificial Intelligence in Statistics, January 1995.  Local copy.

 

·        J. Breese and D. Heckerman.  Decision-theoretic case-based reasoning.  In Proceedings of Fifth Conference on Artificial Intelligence and Statistics, Ft. Lauderdale, FL. Society for Artificial Intelligence in Statistics, January 1995.  Local copy.

 

·        D. Heckerman and R. Shachter.  A decision-based view of causality.  In Proceedings of Fifth Conference on Artificial Intelligence and Statistics, Ft. Lauderdale, FL. Society for Artificial Intelligence in Statistics, January 1995.  Local copy.

 

·        D. Heckerman, J. Breese, K. Rommelse.  Decision-theoretic troubleshootingCACM, 38:49-57, 1995.  Microsoft copy.  Local copy.

 

·        D. Heckerman, M. Wellman.  Bayesian networksCACM, 38:27-30, 1995.

 

·        D. Heckerman, A. Mamdani, M. Wellman.  Real-world applications of Bayesian networks.  CACM, 38:24-26, 1995.

 

·        D. Heckerman and D. Geiger. Learning Bayesian networks: A unification for discrete and Gaussian domains.  In Proceedings of Eleventh Conference on Uncertainty in Artificial Intelligence, Montreal, Quebec, pages 274-284. Morgan Kaufmann, August 1995.  Microsoft copy.  Local copy.

 

·        D. Heckerman, J. Breese, and K. Rommelse.  Troubleshooting under uncertainty.  In Proceedings of Fifth International Workshop on Principles of Diagnosis, New Paltz, NY, pages 121-130, October 1994.  Microsoft copy. 

 

      This paper is the first of a series developing the theory for decision-theoretic troubleshooting, which shipped in early versions of Windows.  Key ingredients include the use of causal models and decision theory.

 

·        D. Heckerman, D. Geiger, and D. Chickering. Learning Bayesian networks: The combination of knowledge and statistical data.  In Proceedings of Tenth Conference on Uncertainty in Artificial  Intelligence, Seattle, WA, pages 293-301. Morgan Kaufmann, July 1994.  Local copy.

 

·        D. Heckerman.  A Tutorial on Learning with Bayesian Networks.  In Learning in Graphical Models, M. Jordan, ed.. MIT Press, Cambridge, MA, 1999.  Also appears as Technical Report MSR-TR-95-06, Microsoft Research, March, 1995.  A similar manuscript appears as Bayesian Networks for Data Mining, Data Mining and Knowledge Discovery, 1: 79-119, 1997.

 

      The 2020 revision corrects errors in the sections on causal learning.

 

·        D. Geiger and D. Heckerman. Learning Gaussian networks.  In Proceedings of Tenth Conference on Uncertainty in Artificial Intelligence, Seattle, WA, pages 235-243. Morgan Kaufmann, July 1994.

 

      This is the first paper on learning Gaussian networks. The latest is here.

     

·        D. Heckerman and J. Breese.  A new look at causal independence.  In Proceedings of Tenth Conference on Uncertainty in Artificial  Intelligence, Seattle, WA, pages 286-292. Morgan Kaufmann, July 1994.  Microsoft copy.  Local copy.

 

·        D. Heckerman and R. Shachter.  A decision-based view of causality.  In Proceedings of Tenth Conference on Uncertainty in Artificial Intelligence, Seattle, WA, pages 302-310. Morgan Kaufmann, July 1994.  Microsoft copy.  Local copy.

 

·        D. Geiger and D. Heckerman.  Inference algorithms for similarity networks.  In Proceedings of Ninth Conference on Uncertainty in Artificial Intelligence, Washington, DC, pages 326-334. Morgan Kaufmann, July 1993.  Microsoft copy.  Local copy.

 

·        D. Heckerman.  Causal independence for knowledge acquisition and inference.  In Proceedings of Ninth Conference on Uncertainty in Artificial Intelligence, Washington, DC, pages 122-127. Morgan Kaufmann, July 1993.  Microsoft copy.  Local copy.

 

·        D. Heckerman and M. Shwe.  Diagnosis of multiple faults: A sensitivity analysis.  In Proceedings of Ninth Conference on Uncertainty in Artificial Intelligence, Washington, DC, pages 80-87. Morgan Kaufmann, July 1993.  Microsoft copy.  Local copy.

 

·        D. Heckerman and E. Horvitz.  Problem formulation as the reduction of a decision model.  In Proceedings of Ninth Conference on Uncertainty in Artificial Intelligence, Washington, DC, pages 80-87. Morgan Kaufmann, July 1993.  Microsoft copy.  Local copy.

 

·        D. Heckerman, E. Horvitz, and B. Middleton.  An approximate nonmyopic computation for value of informationIEEE Transactions on Pattern Analysis and Machine Intelligence, 15:292-298, 1993.  Local copy.

 

·        D. Heckerman, E. Horvitz, and B. Nathwani.  Toward normative expert systems: Part I.  The Pathfinder project.  Methods of Information in Medicine, 31:90-105, 1992.  Microsoft copy.  Semantic Scholar copy.  Local copy.

 

·        D. Heckerman and B. Nathwani.  Toward normative expert systems: Part II. Probability-based representations for efficient knowledge acquisition and inferenceMethods of Information in Medicine, 31:106-116, 1992.  Also in J. van Bemmel and A. McCray, editors, Yearbook of Medical Informatics, pages 430-440. International Medical Informatics Association, Rotterdam, The Netherlands, 1993.  Microsoft copy.  Semantic Scholar copy.  Local copy.

 

·        D. Geiger and D. Heckerman.  Dependence and Relevance: A probabilistic view.  arXiv:1611.02126, Feb 1993.  Local copy.

      In comparison to Chickering, Heckerman, and Meek, JMLR 2004, this paper starts with weaker preconditions (joint distribution is strictly positive binary) and proves a weaker result (total independence implies total disconnectedness).

 

·        D. Heckerman.  The certainty-factor model.  In S. Shapiro, editor, Encyclopedia of Artificial Intelligence, Second Edition, pages 131-138.  Wiley, New York, 1992.  Microsoft copy.  Local copy.

 

·        D. Heckerman and E. Shortliffe.  From certainty factors to belief networks.  Artificial Intelligence in Medicine, 4:35-52, 1992.  Microsoft copy.  Local copy.

 

·        D. Heckerman and B. Nathwani.  An evaluation of the diagnostic accuracy of PathfinderComputers and Biomedical Research, 25:56-74, 1992.  Microsoft copy.  Local copy.

 

·        M. Shwe, B. Middleton, D. Heckerman, M. Henrion, E. Horvitz, H. Lehmann, and G. Cooper.  Probabilistic diagnosis using a reformulation of the INTERNIST-1/QMR knowledge base:  Part 1. The probabilistic model and inference algorithmsMethods of Information in Medicine, 30:241-255, 1991.  Microsoft copy.  Local copy.

 

·        B. Middleton, M. Shwe, D. Heckerman, M. Henrion, E. Horvitz, H. Lehmann, and G. Cooper.  Probabilistic diagnosis using a reformulation of the INTERNIST-1/QMR knowledge base:  Part 2.  Evaluation of diagnostic performanceMethods of Information in Medicine, 30:256-267, 1991.  Microsoft copy.  Local copy.  

 

·        D. Heckerman, E. Horvitz, and B. Middleton.  An approximate nonmyopic computation for value of information.  In Proceedings of the Seventh Conference on Uncertainty in Artificial Intelligence, Los Angeles, CA, pages 135-141. Morgan Kaufmann, July 1991.  Microsoft copy.  Local copy.

 

·         D. Geiger and D. Heckerman.  Advances in probabilistic reasoning.  In Proceedings of the Seventh Conference on Uncertainty in Artificial Intelligence, Los Angeles, CA, pages 118-126. Morgan Kaufmann, July 1991.  Microsoft copy.  Local copy.

 

·       D. Heckerman.  Probabilistic Similarity Networks.  MIT Press, Cambridge, MA, 1991.  Local copy.  Review.

 

      This is a book based on my PhD dissertation.  The technology developed here enabled the practical development of expert systems for diagnosis based on (Bayesian) probabilistic graphical models.  Two companies, Intellipath and Knowledge Industries, were built around this technology, and Microsoft incorporated it to construct the Answer Wizard and Windows Troubleshooters.  The work contains theoretical foundations for probabilistic expert systems, the description of software for building such systems, the description of a working system—Pathfinder—built with this technology, and an evaluation of the diagnostic accuracy of this expert system.

 

·        D. Heckerman.  Probabilistic similarity networksNetworks, 20:607-636, 1990.

 

·        B. Nathwani, D. Heckerman, E. Horvitz, and T. Lincoln.  Integrated expert systems and videodisc in surgical pathology: An overview. Human Pathology, 21:11-27, 1990.

 

·        D. Heckerman.  Similarity networks for the construction of multiple-fault belief  networks.  In Proceedings of the Sixth Conference on Uncertainty in Artificial Intelligence, Boston, MA, pages 32-39. Association for Uncertainty in Artificial Intelligence, Mountain View, CA, July 1990.  Also in P. Bonissone, M. Henrion, L. Kanal, and J. Lemmer, editors, Uncertainty in Artificial Intelligence 6, pages 51-64.  North-Holland, New York, 1990.  Microsoft copy.  Local copy.

 

·        D. Heckerman and E. Horvitz.  Problem formulation as the reduction of a decision model.  In Proceedings of the Sixth Conference on Uncertainty in Artificial Intelligence, Boston, MA, pages 82-89. Association for Uncertainty in Artificial Intelligence, Mountain View, CA, July 1990.  Also in P. Bonissone, M. Henrion, L. Kanal, and J. Lemmer, editors, Uncertainty in Artificial Intelligence 6, pages 159-170.  North-Holland, New York, 1990.  Microsoft copy.  Local copy.

 

·        D. Geiger and D. Heckerman.  Separable and transitive graphoids.  In Proceedings of the Sixth Conference on Uncertainty in Artificial Intelligence, Boston, MA, pages 538-545. Association for Uncertainty in Artificial Intelligence, Mountain View, CA, July 1990.  Microsoft copy.  Local copy.

 

·        H. Suermondt, G. Cooper, and D. Heckerman.  A combination of cutset conditioning with clique-tree propagation in the Pathfinder system.  In Proceedings of the Sixth Conference on Uncertainty in Artificial Intelligence, Boston, MA, pages 273-279. Association for Uncertainty in Artificial Intelligence, Mountain View, CA, July 1990.  Microsoft copy.  Local copy.

 

·        D. Heckerman.  A tractable algorithm for diagnosing multiple diseases.  In Proceedings of the Fifth Workshop on Uncertainty in Artificial Intelligence, Windsor, ON, pages 174-181. Association for Uncertainty in Artificial Intelligence, Mountain View, CA, August 1989.  Also in M. Henrion, R. Shachter, L. Kanal, and J. Lemmer, editors, Uncertainty in Artificial Intelligence 5, pages 163-171.  North-Holland, New York, 1990.  Microsoft copy.  Local copy.

 

·        D. Heckerman, J. Breese, and E. Horvitz.  The compilation of decision models.  In Proceedings of Fifth Workshop on Uncertainty in Artificial Intelligence, Windsor, ON, pages 162-173. Association for Uncertainty in Artificial Intelligence, Mountain View, CA, August 1989.  Microsoft copy.  Local copy.

 

·        D. Heckerman, E. Horvitz, and B. Nathwani.  Update on the Pathfinder project.  In Proceedings of the Thirteenth Symposium on Computer Applications in Medical Care, Washington, D, pages 203-207.  IEEE Computer Society Press, Silver Spring, MD, November 1989.  Microsoft copy.  Local copy.

 

·        E. Horvitz, G. Cooper, and D. Heckerman.  Reflection and action under scarce resources: Theoretical principles and empirical study.  In Proceedings of the Eleventh International Joint Conference on Artificial Intelligence, Detroit, MI, pages 1121-1127. Morgan Kaufmann, San Mateo, CA, August 1989.  Microsoft copy.  Local copy.

 

·        E. Horvitz, D. Heckerman, K. Ng, and B. Nathwani.  Heuristic abstraction in the decision-theoretic Pathfinder system.  In Proceedings of the Thirteenth Symposium on Computer Applications in Medical Care, Washington, DC, pages 178-182. IEEE Computer Society Press, Silver Spring, MD, November 1989.  Microsoft copy.  Local copy.

 

·        D. Heckerman.  An empirical comparison of three inference methods.  In Proceedings of the Fourth Workshop on Uncertainty in Artificial Intelligence, Minneapolis, MN, pages 158-169. Association for Uncertainty in Artificial Intelligence, Mountain View, CA, August 1988.  Also in R. Shachter, T. Levitt, L. Kanal, and J. Lemmer, editors, Uncertainty in Artificial Intelligence 4, pages 283-302.  North-Holland, New York, 1990.  Microsoft copy.  Local copy.

 

·        D. Heckerman and E. Horvitz.  On the expressiveness of rule-based systems for reasoning under uncertainty.  In Proceedings AAAI-87 Sixth National Conference on Artificial Intelligence, Seattle, WA, pages 121-126. Morgan Kaufmann, San Mateo, CA, July 1987.  Local copy.

 

·        D. Heckerman and H. Jimison.  A perspective on confidence and its use in focusing attention during knowledge acquisition.  In Proceedings of the Third Workshop on Uncertainty in Artificial Intelligence, Seattle, WA, pages 123-131. Association for Uncertainty in Artificial Intelligence, Mountain View, CA, July 1987.  Also in L. Kanal, T. Levitt, and J. Lemmer, editors, Uncertainty in Artificial Intelligence 3, pages 123-131. North-Holland, New York, 1989.  Microsoft copy.  Local copy.

 

·        R. Shachter and D. Heckerman.  Thinking backward for knowledge acquisition.   AI Magazine, 8:55-63, 1987.  Microsoft copy.  Local copy.

 

·        D. Heckerman.  An axiomatic framework for belief updates.  In Proceedings of the Second Workshop on Uncertainty in Artificial Intelligence, Philadelphia, PA, pages 123-128. Association for Uncertainty in Artificial Intelligence, Mountain View, CA, August 1986.  Also in L. Kanal and J. Lemmer, editors, Uncertainty in Artificial Intelligence 2, pages 11-22. North-Holland, New York, 1988.  Microsoft copyLocal copy.

 

·        D. Heckerman and E. Horvitz.  The myth of modularity in rule-based systems.  In Proceedings of the Second Workshop on Uncertainty in Artificial Intelligence, Philadelphia, PA, pages 115-121. Association for Uncertainty in Artificial Intelligence, Mountain View, CA, August 1986.  Also in L. Kanal and J. Lemmer, editors, Uncertainty in Artificial Intelligence 2, pages 23-34. North-Holland, New York, 1988.  Microsoft copy.  Local copy.

 

·        D. Heckerman and R. Miller.  Towards a better understanding of the INTERNIST-1 knowledge base.  In Proceedings of Medinfo, Washington, DC, pages 27-31.  North-Holland, New York, October 1986.  Microsoft copy.  Local copy.

 

·        E. Horvitz, D. Heckerman, and C. Langlotz.  A framework for comparing alternative formalisms for plausible reasoning.  In Proceedings AAAI-86 Fifth National Conference on Artificial Intelligence, Philadelphia, PA, pages 210-214. Morgan Kaufmann, San Mateo, CA, August 1986.  Microsoft copy.  Local copy.

 

·        E. Horvitz, D. Heckerman, B. Nathwani, and L. Fagan.  The use of a heuristic problem-solving hierarchy to facilitate the explanation of hypothesis-directed reasoning.  In Proceedings of Medinfo, Washington, DC, pages 27-31.  North-Holland, New York, October 1986.  Microsoft copy.  Local copy.

 

·        E. Horvitz and D. Heckerman.  The inconsistent use of measures of certainty in artificial intelligence research.  In Proceedings of the Workshop on Uncertainty and Probability in Artificial Intelligence, Los Angeles, CA. Association for Uncertainty in Artificial Intelligence, Mountain View, CA, August 1985.  Also in L. Kanal and J. Lemmer, editors, Uncertainty in Artificial Intelligence, pages 137-151. North-Holland, New York, 1986.  Microsoft copy.  Local copy.

 

·        D. Heckerman.  Probabilistic interpretations for MYCIN's certainty factors.  In Proceedings of the Workshop on Uncertainty and Probability in Artificial Intelligence, Los Angeles, CA, pages 9-20. Association for Uncertainty in Artificial Intelligence, Mountain View, CA, August 1985.  Also in L. Kanal. and J. Lemmer, editors, Uncertainty in Artificial Intelligence, pages 167-196. North-Holland, New York, 1986.  Microsoft copy.  Local copy.

 

      Historical note: When this paper was written, the AI community generally believed that probability theory was not suited for encoding uncertainty in expert systems, citing the need to invoke unrealistic assumptions of independence.  In this work, I proved that the certainty-factor model, an alternative framework for uncertain reasoning that the community believed to be superior to probability theory, was in fact equivalent to probability theory with implicit independence assumptions.  I have been told that this work helped move the AI community back to the use of (Bayesian) probabilities for representing uncertainty.

 

·        D. Heckerman, R. Rosenbaum, S. Putterman, and G. Williams.  Pressure release superleak sound modes in He IIJournal of Low Temperature Physics, 38:629, 1980.

 

·        D. Heckerman, S. Garrett, G.A. Williams, and P. Weidman.  Surface tension restoring forces on gravity waves in narrow channelsPhysics of Fluids, 22, 1979.

 

·        S. Putterman, D. Heckerman, R. Rosenbaum, and G. Williams.  Superfluid two-phase soundPhysics Review Letters, 42:580, 1979.

 

·        R. Rosenbaum, G. Williams, D. Heckerman, J. Marcus, D. Scholler, J. Maynard, and I. Rudnick.  Surface tension sound in superfluid helium films adsorbed on alumina powderJournal of Low Temperature Physics, 37:663, 1979.