pith. sign in

arxiv: 2406.05262 · v2 · submitted 2024-06-07 · 📊 stat.AP

A Three-groups Non-local Model for Combining Heterogeneous Data Sources to Identify Genes Associated with Parkinson's Disease

Troy P. Wixson , Benjamin A. Shaby , Daisy L. Philtron , International Parkinson Disease Genomics Consortium , Leandro A. Lima , Stacia K. Wyman , Julia A. Kaye , Steven Finkbeiner This is my paper

Pith reviewed 2026-05-23 23:51 UTC · model grok-4.3

classification 📊 stat.AP
keywords Parkinson's diseasegene identificationmixture modelmulti-omics integrationGWASRNA-seqthree-group modelstatistical genetics
0
0 comments X

The pith

A hierarchical three-group mixture model combines any number of data modalities such as GWAS and RNA-seq into one probability model to identify genes linked to Parkinson's disease.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proposes a statistical framework built around a hierarchical three-group mixture where each gene is probabilistically assigned to a null, deleterious, or beneficial category. Prior group probabilities are drawn from a Dirichlet distribution so that the posterior automatically adjusts for testing thousands of genes at once. Experimental outcomes from different sources are then modeled conditionally on these group labels, which lets the method pool information across heterogeneous data types in a single coherent model. This produces parsimonious inference that controls false positives while increasing power to detect true signals. Simulations show performance comparable to or better than standard GWAS and RNA-seq tools, and the method applied to real data yields novel candidate genes.

Core claim

The central claim is that the three-group formalism, by apportioning prior probability via a Dirichlet distribution and by specifying conditional distributions for each experiment type given the group label, permits any number of data modalities to be combined inside one probability model; the resulting posterior group probabilities deliver both automatic multiplicity correction and information sharing that together yield fewer false positives and greater detection power than separate analyses.

What carries the argument

The hierarchical three-group mixture model with Dirichlet priors on group membership probabilities and conditional likelihoods for each data modality given the group label.

If this is right

  • Additional data modalities can be added by specifying only their conditional distributions given the three group labels.
  • Posterior probabilities for each gene being deleterious or beneficial serve as direct, multiplicity-adjusted measures of evidence.
  • The same model structure can be reused for other complex traits by swapping in the appropriate experiment types.
  • Simulations confirm that the shared information improves power relative to analyzing each modality in isolation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The conditional-modeling approach could be tested on other multi-omics problems such as combining proteomics with methylation data.
  • If the three-group assumption is relaxed to more categories, the Dirichlet construction would still provide automatic multiplicity control.
  • The framework suggests that public repositories of GWAS and expression data could be jointly re-analyzed for many diseases with minimal new experimental cost.

Load-bearing premise

The three-group labels and the conditional distributions of each data type given those labels correctly capture the relevant biology.

What would settle it

A dataset containing known Parkinson's genes in which the model either misses a substantial fraction of them or returns many genes later shown by orthogonal experiments to have no association.

Figures

Figures reproduced from arXiv: 2406.05262 by Benjamin A. Shaby, Daisy L. Philtron, International Parkinson Disease Genomics Consortium, Julia A. Kaye, Leandro A. Lima, Stacia K. Wyman, Steven Finkbeiner, Troy P. Wixson.

Figure 1
Figure 1. Figure 1: The log prior probability mass function for models [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Posterior probability of inclusion in the null group for GWAS only methods, from a single simulated dataset. [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Panel (a) shows posterior probabilities of inclusion in the null group for RNA-seq only methods, from a single [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Boxplots of logarithmic scores (a), Brier scores (b), area under receiver operating characteristic curve (c) [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Boxplots of logarithmic scores (a), Brier scores (b), area under receiver operating characteristic curve (c) [PITH_FULL_IMAGE:figures/full_fig_p017_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Volcano plots for the local model (top row) and piMOM model (bottom row). Effect sizes are the marginal [PITH_FULL_IMAGE:figures/full_fig_p018_6.png] view at source ↗
read the original abstract

We seek to identify genes involved in Parkinson's Disease (PD) by combining information across different experiment types. Each experiment, taken individually, may contain too little information to distinguish some important genes from incidental ones. However, when experiments are combined using the proposed statistical framework, additional power emerges. The fundamental building block of the family of statistical models that we propose is a hierarchical three-group mixture of distributions. Each gene is modeled probabilistically as belonging to either a null group that is unassociated with PD, a deleterious group, or a beneficial group. This three-group formalism has two key features. By apportioning prior probability of group assignments with a Dirichlet distribution, the resultant posterior group probabilities automatically account for the multiplicity inherent in analyzing many genes simultaneously. By building models for experimental outcomes conditionally on the group labels, any number of data modalities may be combined in a single coherent probability model, allowing information sharing across experiment types. These two features result in parsimonious inference with few false positives, while simultaneously enhancing power to detect signals. Simulations show that our three-groups approach performs at least as well as commonly-used tools for GWAS and RNA-seq, and in some cases it performs better. We apply our proposed approach to publicly-available GWAS and RNA-seq datasets, discovering novel genes that are potential therapeutic targets.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript proposes a hierarchical three-group mixture model (null, deleterious, beneficial) with Dirichlet prior on group probabilities for integrating heterogeneous data modalities such as GWAS and RNA-seq to identify Parkinson's Disease associated genes. By modeling experimental outcomes conditionally on latent group labels, the framework enables information sharing across experiment types while the Dirichlet prior provides automatic multiplicity correction. Simulations are reported to show performance at least as good as standard tools, with a real-data application yielding novel candidate genes.

Significance. If the modeling assumptions hold, the approach supplies a coherent joint probability model for multi-modal integration that automatically handles multiple testing and can improve power for gene discovery. The explicit use of simulations benchmarked against common GWAS and RNA-seq tools plus a real-data application constitute concrete strengths that allow direct assessment of operating characteristics.

major comments (2)
  1. [Abstract and Methods] Abstract and Methods: the central claim that conditioning on the three-group labels 'result[s] in parsimonious inference with few false positives, while simultaneously enhancing power' is load-bearing on the assumption that the three discrete groups exhaust the relevant biology and that the chosen conditional distributions p(data|group) for each modality are correctly specified; the manuscript provides no model diagnostics, posterior predictive checks, or sensitivity analyses to this assumption for PD biology.
  2. [Simulations] Simulations section: performance comparisons to standard tools are presented, but without reporting whether the simulated data-generating processes match the model's assumed conditional distributions (or quantifying mismatch), it is unclear whether reported gains in power and false-positive control are attributable to the three-group structure or to favorable simulation design.
minor comments (2)
  1. [Title] Title uses 'non-local model' while the abstract describes a standard hierarchical mixture; clarify whether 'non-local' refers to a specific prior construction or is used in a different sense.
  2. [Abstract] The abstract states that the Dirichlet prior 'automatically account[s] for the multiplicity'; a brief derivation or reference to the implied false-discovery-rate control would strengthen the claim.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments on our manuscript. We address each major comment below and agree that the points raised identify areas where the manuscript can be strengthened through additional clarification and analyses.

read point-by-point responses

  1. Referee: [Abstract and Methods] Abstract and Methods: the central claim that conditioning on the three-group labels 'result[s] in parsimonious inference with few false positives, while simultaneously enhancing power' is load-bearing on the assumption that the three discrete groups exhaust the relevant biology and that the chosen conditional distributions p(data|group) for each modality are correctly specified; the manuscript provides no model diagnostics, posterior predictive checks, or sensitivity analyses to this assumption for PD biology.

    Authors: We agree that the validity of the central claim depends on the three-group structure and conditional distributions being reasonable approximations to the underlying biology. The groups (null, deleterious, beneficial) are chosen to reflect the primary modes of gene association with PD, and the conditional distributions follow standard parametric forms used in GWAS (effect-size or z-score models) and RNA-seq (count-based models). The submitted manuscript does not contain formal posterior predictive checks or sensitivity analyses. We will add these in revision, including sensitivity to the Dirichlet hyperparameters and to alternative conditional distributions, together with posterior predictive diagnostics on both simulated and real data. revision: yes

  2. Referee: [Simulations] Simulations section: performance comparisons to standard tools are presented, but without reporting whether the simulated data-generating processes match the model's assumed conditional distributions (or quantifying mismatch), it is unclear whether reported gains in power and false-positive control are attributable to the three-group structure or to favorable simulation design.

    Authors: The simulations were generated under a range of scenarios intended to include both close alignment with the model's conditional distributions and moderate misspecification. The manuscript, however, does not explicitly report the degree of match or mismatch between the simulation data-generating processes and the model's assumptions. We will revise the Simulations section to describe the data-generation mechanism in detail, quantify the alignment (or deviation) for each scenario, and present performance results stratified by the degree of model match. revision: yes

Circularity Check

0 steps flagged

No circularity: model is a standard hierarchical construction with independent simulation validation

full rationale

The paper defines a new hierarchical three-group mixture model with Dirichlet prior on group probabilities and modality-specific conditional distributions given latent group labels. This structure is constructed directly from the modeling assumptions rather than derived from or reduced to any fitted input or self-citation. Simulations and real-data application are presented as external checks, not as part of the derivation chain. No equations reduce by construction to their inputs, and no load-bearing self-citations or ansatzes are invoked. The derivation is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

2 free parameters · 3 axioms · 0 invented entities

The central claim rests on the validity of the three-group classification, the Dirichlet prior for group probabilities, and the ability to specify appropriate conditional distributions for each data modality.

free parameters (2)
  • Dirichlet concentration parameters
    Hyperparameters controlling the prior probabilities of the three groups across genes.
  • Parameters of conditional distributions
    Parameters defining the outcome distributions for GWAS, RNA-seq, and other modalities given each group label.
axioms (3)
  • domain assumption Genes belong to exactly one of three groups: null, deleterious, or beneficial
    Core modeling choice that partitions the gene space.
  • domain assumption Dirichlet distribution appropriately apportions prior group probabilities to handle multiplicity
    Invoked to automatically account for testing many genes simultaneously.
  • domain assumption Conditional distributions for experimental outcomes given group labels can be specified for any data modality
    Enables the information-sharing mechanism across data types.

pith-pipeline@v0.9.0 · 5797 in / 1323 out tokens · 24451 ms · 2026-05-23T23:51:30.260646+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

63 extracted references · 63 canonical work pages

  1. [1]

    M., Monzón-Sandoval, J., Bowden, R., Alegre-Abarrategui, J., Wade-Martins, R., and Webber, C

    Agarwal, D., Sandor, C., Volpato, V., Caffrey, T. M., Monzón-Sandoval, J., Bowden, R., Alegre-Abarrategui, J., Wade-Martins, R., and Webber, C. (2020). A single-cell atlas of the human substantia nigra reveals cell-specific pathways associated with neurological disorders. Nature communications , 11(1):4183--4183

    work page 2020

  2. [2]

    Albert-Gasc \'o , H., Ros-Bernal, F., Castillo-G \'o mez, E., and Olucha-Bordonau, F. E. (2020). MAP/ERK signaling in developing cognitive and emotional function and its effect on pathological and neurodegenerative processes. International journal of molecular sciences , 21(12):4471

    work page 2020

  3. [3]

    L., Day-Williams, A

    Asimit, J. L., Day-Williams, A. G., Morris, A. P., and Zeggini, E. (2012). ARIEL and AMELIA : testing for an accumulation of rare variants using next-generation sequencing data. Human Heredity , 73(2):84--94

    work page 2012

  4. [4]

    Barbieri, M. M. and Berger, J. O. (2004). Optimal predictive model selection. The Annals of statistics , 32(3):870--897

    work page 2004

  5. [5]

    and Heller, R

    Benjamini, Y. and Heller, R. (2008). Screening for partial conjunction hypotheses. Biometrics , 64(4):1215--1222

    work page 2008

  6. [6]

    and Hochberg, Y

    Benjamini, Y. and Hochberg, Y. (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society Series B-Statistical Methodology , 57(1):289--300

    work page 1995

  7. [7]

    and Beal, M

    Bose, A. and Beal, M. F. (2019). Mitochondrial dysfunction and oxidative stress in induced pluripotent stem cell models of P arkinson's disease. European Journal of Neuroscience , 49(4):525--532

    work page 2019

  8. [8]

    A., Li, Y

    Boyle, E. A., Li, Y. I., and Pritchard, J. K. (2017). An expanded view of complex traits: From polygenic to omnigenic. Cell , 169(7):1177--1186

    work page 2017

  9. [9]

    Brehm, N., Bez, F., Carlsson, T., Kern, B., Gispert, S., Auburger, G., and Cenci, M. (2015). A genetic mouse model of P arkinson’s disease shows involuntary movements and increased postsynaptic sensitivity to apomorphine. Molecular neurobiology , 52:1152--1164

    work page 2015

  10. [10]

    Chang, M., Zhang, Y., Hui, Z., Wang, D., and Guo, H. (2020). IFRD1 regulates the asthmatic responses of airway via NF- B pathway. Molecular Immunology , 127:186--192

    work page 2020

  11. [11]

    Clayton, D. F. and George, J. M. (1998). The synucleins: a family of proteins involved in synaptic function, plasticity, neurodegeneration and disease. Trends in neurosciences , 21(6):249--254

    work page 1998

  12. [12]

    de Valpine , P., Paciorek, C., Turek, D., Michaud, N., Anderson-Bergman, C., Obermeyer, F., Wehrhahn Cortes , C., Rodrìguez, A., Temple Lang , D., and Paganin, S. (2023). NIMBLE : MCMC , Particle Filtering, and Programmable Hierarchical Modeling . R package version 1.0.1

    work page 2023

  13. [13]

    T., and Bodik, R

    de Valpine , P., Turek, D., Paciorek, C., Anderson-Bergman, C., Lang , D. T., and Bodik, R. (2017). Programming with models: writing statistical algorithms for general model structures with NIMBLE . Journal of Computational and Graphical Statistics , 26:403--413

    work page 2017

  14. [14]

    Y., Li, S., Narasimhan, B., and Tibshirani, R

    Ding, D. Y., Li, S., Narasimhan, B., and Tibshirani, R. (2022). Cooperative learning for multiview analysis. Proceedings of the National Academy of Sciences , 119(38):e2202113119

    work page 2022

  15. [15]

    D., and Tusher, V

    Efron, B., Tibshirani, R., Storey, J. D., and Tusher, V. (2001). Empirical bayes analysis of a microarray experiment. Journal of the American Statistical Association , 96(456):1151--1160

    work page 2001

  16. [16]

    Esposito, G., Ana Clara, F., and Verstreken, P. (2012). Synaptic vesicle trafficking and P arkinson's disease. Developmental neurobiology , 72(1):134--144

    work page 2012

  17. [17]

    Fisher, R. (1929). Tests of significance in harmonic analysis. Proceedings of the Royal Society of London Series A-Mathematical and Physical Sciences , 125(796):54--59

    work page 1929

  18. [18]

    K., Nasser, J., Jagadeesh, K

    Gazal, S., Weissbrod, O., Hormozdiari, F., Dey, K. K., Nasser, J., Jagadeesh, K. A., Weiner, D. J., Shi, H., Fulco, C. P., O’Connor, L. J., et al. (2022). Combining snp-to-gene linking strategies to identify disease genes and assess disease omnigenicity. Nature genetics , 54(6):827--836

    work page 2022

  19. [19]

    Gelman, A. (2014). Bayesian data analysis . YBP Print DDA. CRC Press, Boca Raton, third edition

    work page 2014

  20. [20]

    and Wasserman, L

    Genovese, C. and Wasserman, L. (2004). A stochastic process approach to false discovery control. The Annals of Statistics , 32(3):1035--1061

    work page 2004

  21. [21]

    Gerard, D. (2020). Data-based RNA -seq simulations by binomial thinning. BMC bioinformatics , 21(1):206--206

    work page 2020

  22. [22]

    and Raftery, A

    Gneiting, T. and Raftery, A. E. (2007). Strictly proper scoring rules, prediction, and estimation. Journal of the American Statistical Association , 102(477):359--378

    work page 2007

  23. [23]

    Green, P. J. (1995). Reversible jump M arkov chain M onte C arlo computation and B ayesian model determination. Biometrika , 82(4):711--732

    work page 1995

  24. [24]

    R., Dudek, S

    Holzinger, E. R., Dudek, S. M., Frase, A. T., Pendergrass, S. A., and Ritchie, M. D. (2013). Athena: the analysis tool for heritable and environmental network associations. Bioinformatics , 30(5):698--705

    work page 2013

  25. [25]

    Hu, W., Wang, M., Sun, G., Zhang, L., and Lu, H. (2024). Early b cell factor 3 ( EBF3 ) attenuates P arkinson's disease through directly regulating contactin-associated protein-like 4 ( CNTNAP4 ) transcription: An experimental study. Cellular Signalling , pages 111--139

    work page 2024

  26. [26]

    Johnson, V. E. and Rossell, D. (2012). Bayesian model selection in high-dimensional settings. Journal of the American Statistical Association , 107(498):649--660

    work page 2012

  27. [27]

    Johnson, V. E. E. and Rossell, D. (2010). On the use of non-local prior densities in bayesian hypothesis tests. Journal of the Royal Statistical Society. Series B, Statistical methodology , 72(2):143--170

    work page 2010

  28. [28]

    C., and Bernstein, A

    Kochmanski, J., Kuhn, N. C., and Bernstein, A. I. (2022). Parkinson's disease-associated, sex-specific changes in DNA methylation at PARK7 ( DJ -1), SLC17A6 ( VGLUT2 ), PTPRN2 ( IA -2 ), and NR4A2 ( NURR1 ) in cortical neurons. npj Parkinson's Disease , 8(1):120

    work page 2022

  29. [29]

    W., Chen, Y., Shi, W., and Smyth, G

    Law, C. W., Chen, Y., Shi, W., and Smyth, G. K. (2014). voom: precision weights unlock linear model analysis tools for rna-seq read counts. Genome biology , 15(2):R29--R29

    work page 2014

  30. [30]

    and Leal, S

    Li, B. and Leal, S. M. (2008). Methods for detecting associations with rare variants for common diseases: application to analysis of sequence data. The American Journal of Human Genetics , 83(3):311--321

    work page 2008

  31. [31]

    Li, L., Wang, H., Li, H., Lu, X., Gao, Y., and Guo, X. (2022). Long noncoding rna bace1-antisense transcript plays a critical role in P arkinson's disease via microRNA -214-3p/cell death-inducing p53-target protein 1 axis. Bioengineered , 13(4):10889--10901

    work page 2022

  32. [32]

    and Chekouo, T

    Li, W. and Chekouo, T. (2022). Bayesian group selection with non-local priors. Computational statistics , 37(1):287--302

    work page 2022

  33. [33]

    Li, Y., Wu, F.-X., and Ngom, A. (2018). A review on machine learning principles for multi-view biological data integration. Briefings in bioinformatics , 19(2):325--340

    work page 2018

  34. [34]

    W.-L., Leung, C.-T., Pang, S

    Liu, H., Ho, P. W.-L., Leung, C.-T., Pang, S. Y.-Y., Chang, E. E. S., Choi, Z. Y.-K., Kung, M. H.-W., Ramsden, D. B., and Ho, S.-L. (2021). Aberrant mitochondrial morphology and function associated with impaired mitophagy and DNM1L-MAPK/ERK signaling are found in aged mutant P arkinsonian LRRK2R1441G mice. Autophagy , 17(10):3196--3220

    work page 2021

  35. [35]

    B., and Yin, Y

    Liu, Y.-X., Wang, J., Guo, J., Wu, J., Lieberman, H. B., and Yin, Y. (2008). DUSP1 is controlled by p53 during the cellular response to oxidative stress. Molecular Cancer Research , 6(4):624--633

    work page 2008

  36. [36]

    I., Huber, W., and Anders, S

    Love, M. I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and dispersion for rna-seq data with deseq2. Genome biology , 15(12):550--550

    work page 2014

  37. [37]

    Ma, J., Dong, L., Chang, Q., Chen, S., Zheng, J., Li, D., Wu, S., Yang, H., and Li, X. (2023). CXCR4 knockout induces neuropathological changes in the MPTP -lesioned model of P arkinson's disease. Biochimica et Biophysica Acta (BBA)-Molecular Basis of Disease , 1869(2):166597

    work page 2023

  38. [38]

    Madsen, B. E. and Browning, S. R. (2009). A groupwise association test for rare mutations using a weighted sum statistic. PLoS Genetics , 5(2):e1000384

    work page 2009

  39. [39]

    Mochizuki, H., Goto, K., Mori, H., and Mizuno, Y. (1996). Histochemical detection of apoptosis in P arkinson's disease. Journal of the neurological sciences , 137(2):120--123

    work page 1996

  40. [40]

    B., Dermitzakis, E

    Montgomery, S. B., Dermitzakis, E. T., Sammeth, M., Gutierrez-Arcelus, M., Lach, R. P., Ingle, C., Nisbett, J., and Guigo, R. (2010). Transcriptome genetics using second generation sequencing in a caucasian population. Nature (London) , 464(7289):773--777

    work page 2010

  41. [41]

    A., Verstreken, P., Roethig, A., Smet, J., Snellinx, A., Vanbrabant, M., Haddad, D., Frezza, C., Mandemakers, W., Vogt-Weisenhorn, D., et al

    Morais, V. A., Verstreken, P., Roethig, A., Smet, J., Snellinx, A., Vanbrabant, M., Haddad, D., Frezza, C., Mandemakers, W., Vogt-Weisenhorn, D., et al. (2009). Parkinson's disease mutations in PINK1 result in decreased complex I activity and deficient synaptic function. EMBO molecular medicine , 1(2):99--111

    work page 2009

  42. [42]

    and Thilly, W

    Morgenthaler, S. and Thilly, W. G. (2007). A strategy to discover genes that carry multi-allelic or mono-allelic risk for common diseases: a cohort allelic sums test (cast). Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis , 615(1):28--56

    work page 2007

  43. [43]

    A., Pankratz, N., Lill, C

    Nalls, M. A., Pankratz, N., Lill, C. M., Do, C. B., Hernandez, D. G., Saad, M., DeStefano, A. L., Kara, E., Bras, J., Sharma, M., Schulte, C., Keller, M. F., Arepalli, S., Letson, C., Edsall, C., Stefansson, H., Liu, X., Pliner, H., Lee, J. H., Cheng, R., Ikram, M. A., Ioannidis, J. P. A., Hadjigeorgiou, G. M., Bis, J. C., Martinez, M., Perlmutter, J. S.,...

    work page 2014

  44. [44]

    Nikooienejad, A., Wang, W., and Johnson, V. E. (2016). Bayesian variable selection for binary outcomes in high-dimensional genomic studies using non-local priors. BIOINFORMATICS , 32(9):1338--1345

    work page 2016

  45. [45]

    M., Alexopoulou, Z., and Tofaris, G

    Perrett, R. M., Alexopoulou, Z., and Tofaris, G. K. (2015). The endosomal pathway in P arkinson's disease. Molecular and Cellular Neuroscience , 66:21--28

    work page 2015

  46. [46]

    K., Gilad, Y., Pritchard, J

    Pickrell, J. K., Gilad, Y., Pritchard, J. K., Marioni, J. C., Pai, A. A., Degner, J. F., Engelhardt, B. E., Nkadori, E., Veyrieras, J.-B., and Stephens, M. (2010). Understanding mechanisms underlying human gene expression variation with rna sequencing. Nature (London) , 464(7289):768--772

    work page 2010

  47. [47]

    C., and Sun, W

    Richardson, S., Tseng, G. C., and Sun, W. (2016). Statistical methods in integrative genomics. Annual review of statistics and its application , 3:181--209

    work page 2016

  48. [48]

    D., Holzinger, E

    Ritchie, M. D., Holzinger, E. R., Li, R., Pendergrass, S. A., and Kim, D. (2015). Methods of integrating data to uncover genotype-phenotype interactions. Nature Reviews. Genetics , 16(2):85

    work page 2015

  49. [49]

    D., McCarthy, D

    Robinson, M. D., McCarthy, D. J., and Smyth, G. K. (2010). edgeR : a bioconductor package for differential expression analysis of digital gene expression data. BIOINFORMATICS , 26(1):139--140

    work page 2010

  50. [50]

    Schilder, B. M. and Raj, T. (2022). Fine-mapping of P arkinson’s disease susceptibility loci identifies putative causal variants. Human Molecular Genetics , 31(6):888--900

    work page 2022

  51. [51]

    Scott, J. G. and Berger, J. O. (2010). Bayes and empirical- B ayes multiplicity adjustment in the variable-selection problem. The Annals of Statistics. , 38(5):2587--2619

    work page 2010

  52. [52]

    Shin, M., Bhattacharya, A., and Johnson, V. E. (2018). Scalable B ayesian variable selection using nonlocal prior densities in ultrahigh-dimensional settings. Statist. Sinica , 28(2):1053--1078

    work page 2018

  53. [53]

    A., Suchman, E

    Stouffer, S. A., Suchman, E. A., DeVinney, L. C., Star, S. A., and Williams Jr, R. M. (1949). The A merican soldier: A djustment during army life, V ol. I . Studies in Social Psychology World War II

    work page 1949

  54. [54]

    A., Harvey, K., Rossi, L., Ferraina, C., De Biase, V., Rodolfo, C., Harvey, R

    Strobbe, D., Robinson, A. A., Harvey, K., Rossi, L., Ferraina, C., De Biase, V., Rodolfo, C., Harvey, R. J., and Campanella, M. (2018). Distinct mechanisms of pathogenic DJ -1 mutations in mitochondrial quality control. Frontiers in Molecular Neuroscience , 11:68

    work page 2018

  55. [55]

    Tan, E.-K., Chao, Y.-X., West, A., Chan, L.-L., Poewe, W., and Jankovic, J. (2020). Parkinson disease and the immune system—associations, mechanisms and therapeutics. Nature Reviews Neurology , 16(6):303--318

    work page 2020

  56. [56]

    G., Chalmers-Redman, R., Brown, D., and Tatton, N

    Tatton, W. G., Chalmers-Redman, R., Brown, D., and Tatton, N. (2003). Apoptosis in P arkinson's disease: signals for neuronal degradation. Annals of Neurology: Official Journal of the American Neurological Association and the Child Neurology Society , 53(S3):S61--S72

    work page 2003

  57. [57]

    Toskas, K., Yaghmaeian-Salmani, B., Skiteva, O., Paslawski, W., Gillberg, L., Skara, V., Antoniou, I., S \"o dersten, E., Svenningsson, P., Chergui, K., et al. (2022). PRC2 -mediated repression is essential to maintain identity and function of differentiated dopaminergic and serotonergic neurons. Science Advances , 8(34):eabo1543

    work page 2022

  58. [58]

    Tyekucheva, S., Marchionni, L., Karchin, R., and Parmigiani, G. (2011). Integrating diverse genomic data using gene sets. Genome biology , 12(10):R105--R105

    work page 2011

  59. [59]

    Q., Munung, N

    Uffelmann, E., Huang, Q. Q., Munung, N. S., De Vries, J., Okada, Y., Martin, A. R., Martin, H. C., Lappalainen, T., and Posthuma, D. (2021). Genome-wide association studies. Nature Reviews Methods Primers , 1(1):59

    work page 2021

  60. [60]

    Usenko, T., Bezrukova, A., Basharova, K., Panteleeva, A., Nikolaev, M., Kopytova, A., Miliukhina, I., Emelyanov, A., Zakharova, E., and Pchelina, S. (2021). Comparative transcriptome analysis in monocyte-derived macrophages of asymptomatic gba mutation carriers and patients with GBA -associated P arkinson’s disease. Genes , 12(10):1545

    work page 2021

  61. [61]

    J., and Wu, G

    Wang, J., Zhou, J.-Y., Kho, D., Reiners Jr, J. J., and Wu, G. S. (2016). Role for DUSP1 (dual-specificity protein phosphatase 1) in the regulation of autophagy. Autophagy , 12(10):1791--1803

    work page 2016

  62. [62]

    Zaltieri, M., Longhena, F., Pizzi, M., Missale, C., Spano, P., Bellucci, A., et al. (2015). Mitochondrial dysfunction - S ynuclein synaptic pathology in P arkinson’s disease: who’s on first? Parkinson's disease , 2015

    work page 2015

  63. [63]

    Zhao, C., Datta, S., Mandal, P., Xu, S., and Hamilton, T. (2010). Stress-sensitive regulation of IFRD1 mRNA decay is mediated by an upstream open reading frame. Journal of Biological Chemistry , 285(12):8552--8562

    work page 2010