pith. sign in

arxiv: 2606.13556 · v1 · pith:VGR5T2K7new · submitted 2026-06-11 · 💻 cs.AI · cs.HC· q-bio.BM· q-bio.GN· q-bio.MN

Is It You or Your Environment? A Bayesian Inference Framework for Genomically-Anchored Personalized Physiological Interpretation

Aruna Dey , Suraj Biswas This is my paper

Pith reviewed 2026-06-27 06:56 UTC · model grok-4.3

classification 💻 cs.AI cs.HCq-bio.BMq-bio.GNq-bio.MN
keywords genomic anchorBayesian priorphysiological set pointpersonalized healthcold-start problemGWAS effect sizesMendelian randomization
0
0 comments X

The pith

An individual's genomic profile initializes a Bayesian prior over physiological set points that separates constitutional from environmental effects from the first measurement.

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

The paper addresses the cold-start problem in personalized health AI, where models need weeks of individual data to distinguish fixed traits from environmental influences. It proposes treating an individual's genomic profile as an exogenous anchor that sets a personalized Bayesian prior for their physiological set point using GWAS effect sizes. This prior enables calculation of non-constitutional deviation immediately upon the first measurement. The prior then decays over time, shifting weight toward accumulating empirical observations. The framework is outlined across six physiological domains while grading anchors by evidence strength and requiring ancestry-matched effect sizes.

Core claim

An individual's genomic profile serves as an exogenous genetic anchor that initializes a Bayesian belief state over physiological set point G-hat = mu + sum(beta_i * g_i), allowing separation of constitutional from environmental deviation from the first measurement onward.

What carries the argument

The genomic prior G-hat = mu + sum(beta_i * g_i) that initializes the Bayesian state and decays via a time-weighted function toward empirical averages.

If this is right

  • The same observed value, such as HRV of 55 ms, generates a suppression hypothesis for one genomic prior and an enhancement hypothesis for another.
  • The belief state transitions from genome-dominated to data-dominated inference through the decay function G-hat_t = w(t)*G-hat_genomic + [1-w(t)]*P-bar_t.
  • Priors must be evidence-graded, favoring robustly replicated loci over contested candidate genes.
  • Valid deployment requires ancestry-matched effect sizes and produces attributions rather than deterministic outputs.

Where Pith is reading between the lines

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

  • Consumer genetic data could enable immediate baseline separation in health monitoring apps without an initial observation period.
  • The approach could be tested by comparing G-hat predictions against multi-year empirical averages in large cohorts while controlling for environment.
  • Similar anchoring might apply to other domains where stable genetic predictors exist for individual set points.

Load-bearing premise

GWAS-derived effect sizes can be treated as accurate, fixed, individual-level predictors of physiological set points that are immune to reverse causation and transferable across individuals within ancestry groups.

What would settle it

Long-term average physiological measurements for individuals with known genomic profiles that deviate substantially and consistently from their G-hat predictions in patterns not explained by measured environmental or behavioral factors.

Figures

Figures reproduced from arXiv: 2606.13556 by Aruna Dey, Suraj Biswas.

Figure 1
Figure 1. Figure 1: Genotype G is the root node (no incoming edges: it cannot be caused by downstream variables). Environment E and G together produce the phenotype P, which is what the sensor measures. Decomposing the observed signal into the genetic set point Gˆ and the deviation P − Gˆ isolates the non-constitutional, actionable component. Because G is exogenous, the deviation cannot be an artifact of reverse causation fro… view at source ↗
Figure 2
Figure 2. Figure 2: Both persons A and B show an identical observed HRV of 55 ms, which falls within the population norm band (shaded). Against their respective genetic set points, however, Person A (high set point, 80 ms) shows a negative deviation δA = −25 ms suggesting physiological suppression, while Person B (low set point, 30 ms) shows a positive deviation δB = +25 ms suggesting environmental support. Same observed valu… view at source ↗
Figure 3
Figure 3. Figure 3: Panel A: Published per-risk-allele GWAS effect sizes with 95% CI bars for six SNPs across five physiological domains. FTO rs9939609 → BMI: β ≈ 0.36 kg/m2 [10, 28]; FADS1 rs174537 → circulating PUFA ratio: β ≈ 0.30 SD [11]; FKBP5 rs1360780 → cortisol AUC: β ≈ 0.19 SD [12]; GNG11 rs9954924 → RMSSD (HRV): β ≈ 0.08 SD [2]; CRY1 rs2287161 → chronotype: β ≈ 0.06 SD [4]; COMT rs4680 → prefrontal dopamine proxy: β… view at source ↗
Figure 4
Figure 4. Figure 4: Panel A: The rs1360780 T-allele (risk) increases FKBP5 co-chaperone activity, reducing glu￾cocorticoid receptor sensitivity and slowing cortisol feedback termination; the effect is allele-dose-dependent (TT > TC > CC). Early-life adversity amplifies genetic risk effects through ↑ HPA-axis drive, ↑ inflam￾mation, and ↑ allostatic load. Panel B: Schematic cortisol recovery curves following an acute stressor,… view at source ↗
Figure 5
Figure 5. Figure 5: [18]. The genomic-anchor framework cleanly delivers Rung 1: a calibrated, ranked causal hypothesis. Rung 2 and Rung 3 require either population-level Mendelian randomization or individual-level experimental evidence. Observation alone supports attribution; stronger causal claims require intervention or counterfactual evidence. data alone cannot settle. The caveat is horizontal pleiotropy: if the instrument… view at source ↗
Figure 6
Figure 6. Figure 6: Moving from a ranked causal hypothesis to a claim about a specific event requires within-person experimental evidence. The ABAB design—repeated introduction and withdrawal of the candidate cause— provides this [20]. If δ rises when the candidate cause is present and falls when it is removed, token causal support is strengthened. The exogenous genetic anchor improves power by supplying a stable constitution… view at source ↗
Figure 7
Figure 7. Figure 7: The exogenous genetic anchor carries full weight at cold-start (Day 0) and decays as the empirical personal behavioral baseline accumulates, settling at a non-zero floor (≈30%) where it continues as a weak interpretive anchor. The blending formula Gˆ t = w(t)Gˆ genomic+[1−w(t)]P¯ t governs the transition. The exact decay function, time constant, and floor value are implementation decisions; the qualitative… view at source ↗
Figure 8
Figure 8. Figure 8: Each step in the chain from population cohort to genetic set point Gˆ depends on the beta value βi—the per-allele effect size estimated by GWAS. When the beta is derived from a different ancestry than the individual being assessed (the dominant case for South Asian populations), the set point is biased and the deviation δ becomes unreliable as a causal-attribution signal. Correcting this requires ancestry-… view at source ↗
read the original abstract

Personalized health AI systems face a fundamental cold-start problem: machine learning models for physiological interpretation require weeks of individual behavioral data before they can distinguish constitutional variation from environmentally driven deviation. We propose a solution grounded in causal inference and Bayesian prior design. An individual's genomic profile serves as an exogenous genetic anchor -- a domain-informed, personalized prior that is fixed at conception, immune to reverse causation, and available before a single behavioral observation is collected. The anchor initializes a Bayesian belief state over an individual's physiological set point G-hat = mu + sum(beta_i * g_i), where beta_i are GWAS-derived effect sizes and g_i are risk-allele counts. Each incoming physiological measurement P produces a non-constitutional deviation delta = P - G-hat that separates the signal attributable to environment and state from the constitutionally fixed baseline. As behavioral data accrue, the prior decays according to G-hat_t = w(t)*G-hat_genomic + [1-w(t)]*P-bar_t, transitioning from genome-dominated to empirical-baseline-dominated inference. The same observed HRV of 55 ms generates a suppression hypothesis for a person whose prior predicts 80 ms, and an enhancement hypothesis for a person whose prior predicts 30 ms -- a reversal impossible without a personalized anchor. We develop this architecture across six physiological domains, grading genomic priors by evidence strength, distinguishing robustly replicated anchors (FTO, FADS1/2, FKBP5) from contested candidate genes (SLC6A4, MAOA, DRD2). We address the inference boundary between association, Mendelian randomization, and individual token causation, and define four constraints for deployment: evidence-graded priors, dynamic decay, ancestry-matched effect sizes, and attribution rather than deterministic output.

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

3 major / 1 minor

Summary. The paper claims that genomic profiles provide an exogenous, conception-fixed anchor G-hat = mu + sum(beta_i * g_i) (with beta_i from GWAS) that initializes a Bayesian prior over physiological set points, allowing delta = P - G-hat to separate constitutional from environmental effects from the first measurement; a decaying weight w(t) transitions the prior to empirical baselines, producing reversed interpretations of the same observation (e.g., HRV=55 ms) depending on the genomic prior, and the architecture is developed across six domains with evidence grading and Mendelian-randomization boundaries.

Significance. If the genomic anchors deliver usable out-of-sample precision and the separation holds, the approach would address the cold-start problem in personalized physiological AI by supplying individualized baselines before behavioral data arrive, with potential value for traits having robust replicated loci.

major comments (3)
  1. [Abstract] Abstract, definition of G-hat: the quantity is constructed directly from externally fitted population-level GWAS betas and allele counts, so the subsequent claim that delta = P - G-hat isolates non-constitutional deviation reduces by construction to a residual whose value is set by parameters external to the individual's data stream rather than an independent inference.
  2. [Abstract] Abstract, paragraph on reversal example and six domains: no reported R^2, out-of-sample accuracy, ancestry transferability, or sensitivity bounds are supplied for the polygenic scores in any of the six physiological domains, so the practical informativeness of the prior (and thus the claimed reversal of suppression vs. enhancement hypotheses) remains unquantified.
  3. [Abstract] Abstract, discussion of evidence grading and Mendelian randomization: the framework invokes these boundaries but supplies no quantitative threshold on prior precision or variance explained below which the anchor collapses toward the population mean and the cold-start solution becomes ineffective.
minor comments (1)
  1. [Abstract] Notation for w(t) and the transition G-hat_t is introduced without an explicit functional form or example parameterization, making the decay schedule difficult to reproduce or critique.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for these precise comments on the abstract. We respond to each point below, clarifying the intended role of the external genomic anchor and committing to revisions that strengthen the presentation of its limitations and operational boundaries.

read point-by-point responses

  1. Referee: [Abstract] Abstract, definition of G-hat: the quantity is constructed directly from externally fitted population-level GWAS betas and allele counts, so the subsequent claim that delta = P - G-hat isolates non-constitutional deviation reduces by construction to a residual whose value is set by parameters external to the individual's data stream rather than an independent inference.

    Authors: We agree that G-hat is constructed from external GWAS parameters and that delta is therefore a residual relative to those parameters rather than an inference drawn solely from the individual's own data. This external construction is a core design choice: the anchor is intended to be conception-fixed and exogenous, supplying an individualized starting belief before any physiological observations exist. The subsequent Bayesian update then incorporates the individual's data stream. We will revise the abstract to state explicitly that the separation relies on this externally supplied prior and does not constitute an independent inference from the data alone. revision: yes

  2. Referee: [Abstract] Abstract, paragraph on reversal example and six domains: no reported R^2, out-of-sample accuracy, ancestry transferability, or sensitivity bounds are supplied for the polygenic scores in any of the six physiological domains, so the practical informativeness of the prior (and thus the claimed reversal of suppression vs. enhancement hypotheses) remains unquantified.

    Authors: The referee correctly observes that no quantitative metrics (R^2, out-of-sample performance, ancestry transferability, or sensitivity bounds) appear in the abstract or are newly computed in the manuscript. The reversal example is conceptual, intended to show the logical consequence of a personalized prior rather than to demonstrate a specific predictive gain. We will add a table summarizing published variance-explained estimates and replication status for the cited anchors (FTO, FADS1/2, FKBP5 and others) drawn from the GWAS literature, together with brief notes on ancestry considerations. This addition will quantify the expected strength of the priors without introducing new empirical validation. revision: partial

  3. Referee: [Abstract] Abstract, discussion of evidence grading and Mendelian randomization: the framework invokes these boundaries but supplies no quantitative threshold on prior precision or variance explained below which the anchor collapses toward the population mean and the cold-start solution becomes ineffective.

    Authors: We acknowledge that the manuscript does not define an explicit numerical threshold (e.g., minimum variance explained or replication criteria) at which the genomic anchor would be down-weighted to the population mean. The current evidence-grading scheme is qualitative. We will introduce a concrete operational rule in the revised text—for instance, requiring that a polygenic score explain at least 0.5 % of trait variance in a meta-analysis of >50 000 individuals or have been replicated in at least two independent cohorts—below which the weight w(t) on the genomic component is automatically reduced. revision: yes

Circularity Check

1 steps flagged

Genomic set-point subtraction defines environmental deviation by construction

specific steps
  1. self definitional [Abstract]
    "The anchor initializes a Bayesian belief state over an individual's physiological set point G-hat = mu + sum(beta_i * g_i), where beta_i are GWAS-derived effect sizes and g_i are risk-allele counts. Each incoming physiological measurement P produces a non-constitutional deviation delta = P - G-hat that separates the signal attributable to environment and state from the constitutionally fixed baseline."

    G-hat is constructed directly from fitted GWAS parameters; delta is then defined as the arithmetic difference from that constructed quantity. The separation of 'environment and state' is therefore true by the paper's own definitional equation rather than derived from independent evidence or inference steps.

full rationale

The paper's central separation of constitutional from environmental effects is achieved by defining G-hat from external GWAS betas and then setting delta = P - G-hat. This reduction is explicit in the provided equations and does not constitute an independent derivation; the claimed separation follows immediately from the definitional subtraction rather than from additional causal or Bayesian machinery.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 1 invented entities

The framework rests on the transferability of population-level GWAS effect sizes to individuals, the validity of treating genetic variants as exogenous to current physiology, and an unspecified functional form for the decay weight w(t). No independent evidence for these is supplied in the text.

free parameters (2)
  • w(t)
    Time-dependent weight that blends genomic prior with empirical mean; functional form and rate are not specified and must be chosen to make the transition work.
  • beta_i (GWAS effect sizes)
    Taken from external GWAS studies; treated as fixed inputs but are themselves fitted quantities whose applicability to the target individual is assumed.
axioms (2)
  • domain assumption GWAS-derived effect sizes are accurate individual-level predictors of physiological set points and are immune to reverse causation.
    Invoked when defining G-hat = mu + sum(beta_i * g_i) and when claiming the anchor is fixed at conception.
  • domain assumption Ancestry-matched effect sizes can be selected without introducing new bias.
    Listed as one of the four deployment constraints but not derived.
invented entities (1)
  • G-hat (genomically anchored physiological set point) no independent evidence
    purpose: Serves as the personalized Bayesian prior that separates constitutional from environmental deviation.
    New ledger entry constructed from GWAS betas; no independent falsifiable handle supplied beyond the GWAS literature itself.

pith-pipeline@v0.9.1-grok · 5871 in / 1638 out tokens · 15033 ms · 2026-06-27T06:56:41.001047+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

37 extracted references · 34 canonical work pages

  1. [1]

    Predicting personality from patterns of behavior collected withsmartphones.Proceedings of the National Academy of Sciences117(30):17680–17687, 2020

    Stachl C, Au Q, Schoedel R, et al. Predicting personality from patterns of behavior collected withsmartphones.Proceedings of the National Academy of Sciences117(30):17680–17687, 2020. doi:10.1073/pnas.1920484117

    work page doi:10.1073/pnas.1920484117 2020

  2. [2]

    Genetic loci associated with heart rate vari- ability and their effects on cardiac disease risk.Nature Communications8:15805, 2017

    Nolte IM, Munoz ML, Tragante V, et al. Genetic loci associated with heart rate vari- ability and their effects on cardiac disease risk.Nature Communications8:15805, 2017. doi:10.1038/ncomms15805

    work page doi:10.1038/ncomms15805 2017

  3. [3]

    Heritability estimates of the Big Five personality traits based on common genetic variants.Translational Psychiatry5:e604, 2015

    Power RA, Pluess M. Heritability estimates of the Big Five personality traits based on common genetic variants.Translational Psychiatry5:e604, 2015. doi:10.1038/tp.2015.96

    work page doi:10.1038/tp.2015.96 2015

  4. [4]

    Genome-wide association analyses of chronotype in 697,828 individuals provides insights into circadian rhythms.Nature Communications10:343,

    Jones SE, Lane JM, Wood AR, et al. Genome-wide association analyses of chronotype in 697,828 individuals provides insights into circadian rhythms.Nature Communications10:343,

  5. [5]

    doi:10.1038/s41467-018-08259-7

    work page doi:10.1038/s41467-018-08259-7

  6. [6]

    Finding the missing heritability of complex diseases

    Manolio TA, Collins FS, Cox NJ, et al. Finding the missing heritability of complex diseases. Nature461:747–753, 2009. doi:10.1038/nature08494

    work page doi:10.1038/nature08494 2009

  7. [7]

    An expanded view of complex traits: From polygenic to omnigenic.Cell169(7):1177–1186, 2017

    Boyle EA, Li YI, Pritchard JK. An expanded view of complex traits: From polygenic to omnigenic.Cell169(7):1177–1186, 2017. doi:10.1016/j.cell.2017.05.038

    work page doi:10.1016/j.cell.2017.05.038 2017

  8. [8]

    Border R, Johnson EC, Evans LM, et al. No support for historical candidate gene or candidate gene-by-interaction hypotheses for major depression across multiple large samples.American Journal of Psychiatry176(5):376–387, 2019. doi:10.1176/appi.ajp.2018.18070881

    work page doi:10.1176/appi.ajp.2018.18070881 2019

  9. [9]

    Interaction between the serotonin transporter gene (5- HTTLPR), stressful life events, and risk of depression: a meta-analysis.JAMA301(23):2462– 2471, 2009

    Risch N, Herrell R, Lehner T, et al. Interaction between the serotonin transporter gene (5- HTTLPR), stressful life events, and risk of depression: a meta-analysis.JAMA301(23):2462– 2471, 2009. doi:10.1001/jama.2009.878

    work page doi:10.1001/jama.2009.878 2009

  10. [10]

    Collaborative meta-analysis finds no evidence ofastronginteractionbetweenstressand5-HTTLPRgenotypecontributingtothedevelopment of depression.Molecular Psychiatry23(1):133–142, 2018

    Culverhouse RC, Saccone NL, Horton AC, et al. Collaborative meta-analysis finds no evidence ofastronginteractionbetweenstressand5-HTTLPRgenotypecontributingtothedevelopment of depression.Molecular Psychiatry23(1):133–142, 2018. doi:10.1038/mp.2017.44

    work page doi:10.1038/mp.2017.44 2018

  11. [11]

    A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity.Science316(5826):889– 894, 2007

    Frayling TM, Timpson NJ, Weedon MN, et al. A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity.Science316(5826):889– 894, 2007. doi:10.1126/science.1141634

    work page doi:10.1126/science.1141634 2007

  12. [12]

    Genome-wide association study of plasma polyun- saturated fatty acids in the InCHIANTI study.PLoS Genetics5(1):e1000338, 2009

    Tanaka T, Shen J, Abecasis GR, et al. Genome-wide association study of plasma polyun- saturated fatty acids in the InCHIANTI study.PLoS Genetics5(1):e1000338, 2009. doi:10.1371/journal.pgen.1000338

    work page doi:10.1371/journal.pgen.1000338 2009

  13. [13]

    Allele-specific FKBP5 DNA demethylation mediates gene–childhood trauma interactions.Nature Neuroscience16(1):33–41, 2013

    Klengel T, Mehta D, Anacker C, et al. Allele-specific FKBP5 DNA demethylation mediates gene–childhood trauma interactions.Nature Neuroscience16(1):33–41, 2013. doi:10.1038/nn.3275

    work page doi:10.1038/nn.3275 2013

  14. [14]

    Chang HA, Fang WH, Wan FJ, et al. Age-specific associations among functional COMT Val158Met polymorphism, resting parasympathetic nervous control and generalized anxiety disorder.Psychoneuroendocrinology106:58–65, 2019. doi:10.1016/j.psyneuen.2019.03.020. 21

    work page doi:10.1016/j.psyneuen.2019.03.020 2019

  15. [15]

    Effect of COMT Val108/158 Met genotype on frontal lobe function and risk for schizophrenia.Proceedings of the National Academy of Sciences98(12):6917–6922, 2001

    Egan MF, Goldberg TE, Kolachana BS, et al. Effect of COMT Val108/158 Met genotype on frontal lobe function and risk for schizophrenia.Proceedings of the National Academy of Sciences98(12):6917–6922, 2001. doi:10.1073/pnas.111134598

    work page doi:10.1073/pnas.111134598 2001

  16. [16]

    Inverted-U-shaped dopamine actions on human work- ing memory and cognitive control.Biological Psychiatry69(12):e113–e125, 2011

    Cools R, D’Esposito M. Inverted-U-shaped dopamine actions on human work- ing memory and cognitive control.Biological Psychiatry69(12):e113–e125, 2011. doi:10.1016/j.biopsych.2011.03.028

    work page doi:10.1016/j.biopsych.2011.03.028 2011

  17. [17]

    Jung Y, Montel RA, Shen PH, Mash DC, Goldman D. Assessment of the association of D2 dopamine receptor gene and reported allele frequencies with alcohol use disor- ders: a systematic review and meta-analysis.JAMA Network Open2(11):e1914940, 2019. doi:10.1001/jamanetworkopen.2019.14940

    work page doi:10.1001/jamanetworkopen.2019.14940 2019

  18. [18]

    Clinical use of current polygenic risk scores may exacerbate health disparities.Nature Genetics51:584–591, 2019

    Martin AR, Kanai M, Kamatani Y, et al. Clinical use of current polygenic risk scores may exacerbate health disparities.Nature Genetics51:584–591, 2019. doi:10.1038/s41588-019-0379- x

    work page doi:10.1038/s41588-019-0379- 2019

  19. [19]

    New York: Basic Books, 2018

    Pearl J, Mackenzie D.The Book of Why: The New Science of Cause and Effect. New York: Basic Books, 2018

    2018

  20. [20]

    Mendelian randomization: can genetic epidemiology contribute to understanding environmental determinants of disease?International Journal of Epidemiology 32(1):1–22, 2003

    Davey Smith G, Ebrahim S. Mendelian randomization: can genetic epidemiology contribute to understanding environmental determinants of disease?International Journal of Epidemiology 32(1):1–22, 2003. doi:10.1093/ije/dyg070

    work page doi:10.1093/ije/dyg070 2003

  21. [21]

    Personalized medicine: time for one-person trials.Nature520(7549):609–611, 2015

    Schork NJ. Personalized medicine: time for one-person trials.Nature520(7549):609–611, 2015. doi:10.1038/520609a

    work page doi:10.1038/520609a 2015

  22. [22]

    Econometrica 37, 424–438

    Granger CWJ. Investigating causal relations by econometric models and cross-spectral meth- ods.Econometrica37(3):424–438, 1969. doi:10.2307/1912791

    work page doi:10.2307/1912791 1969

  23. [23]

    Harnessing smartphone-based digital phenotyping to enhance behavioral and mental health.Neuropsychopharmacology41(7):1691–1696, 2016

    Onnela JP, Rauch SL. Harnessing smartphone-based digital phenotyping to enhance behavioral and mental health.Neuropsychopharmacology41(7):1691–1696, 2016. doi:10.1038/npp.2016.7

    work page doi:10.1038/npp.2016.7 2016

  24. [24]

    Obesity- associated variants within FTO form long-range functional connections with IRX3.Nature 507(7492):371–375, 2014

    Smemo S, Tena JJ, Kim KH, Gamazon ER, Sakabe NJ, Gómez-Marín C, et al. Obesity- associated variants within FTO form long-range functional connections with IRX3.Nature 507(7492):371–375, 2014. doi:10.1038/nature13138

    work page doi:10.1038/nature13138 2014

  25. [25]

    Genetic influences on heart rate variability.International Journal of Psychophysiology108:95–100, 2016

    Kim J, Woo BH, Chae JH, Kim CK. Genetic influences on heart rate variability.International Journal of Psychophysiology108:95–100, 2016. doi:10.1016/j.ijpsycho.2016.04.010

    work page doi:10.1016/j.ijpsycho.2016.04.010 2016

  26. [26]

    Binder EB. The role of FKBP5, a co-chaperone of the glucocorticoid receptor in the patho- genesis and therapy of affective and anxiety disorders.Psychoneuroendocrinology34(Suppl 1):S186–S195, 2009. doi:10.1016/j.psyneuen.2009.05.021

    work page doi:10.1016/j.psyneuen.2009.05.021 2009

  27. [27]

    Digital biomarkers of mood disorders and symptom change.NPJ Digital Medicine2:3, 2019

    Jacobson NC, Weingarden H, Wilhelm S. Digital biomarkers of mood disorders and symptom change.NPJ Digital Medicine2:3, 2019. doi:10.1038/s41746-019-0078-8

    work page doi:10.1038/s41746-019-0078-8 2019

  28. [28]

    Silventoinen K, Jelenkovic A, Sund R, et al. Differences in genetic and environmental vari- ation in adult BMI by sex, age, time period, and region: an individual-based pooled analysis of 40 twin cohorts.American Journal of Clinical Nutrition106(2):457–466, 2017. doi:10.3945/ajcn.117.153643. 22

    work page doi:10.3945/ajcn.117.153643 2017

  29. [29]

    Association analyses of 249,796 individuals re- veal 18 new loci associated with body mass index.Nature Genetics42(11):937–948, 2010

    Speliotes EK, Willer CJ, Berndt SI, et al. Association analyses of 249,796 individuals re- veal 18 new loci associated with body mass index.Nature Genetics42(11):937–948, 2010. doi:10.1038/ng.686

    work page doi:10.1038/ng.686 2010

  30. [30]

    Omega-3 fatty acids and inflammatory processes.Nutrients2(3):355–374, 2010

    Calder PC. Omega-3 fatty acids and inflammatory processes.Nutrients2(3):355–374, 2010. doi:10.3390/nu2030355

    work page doi:10.3390/nu2030355 2010

  31. [31]

    The FADS1 genotypes modify the effect of linoleic acid-enriched diet on adipose tissue inflammation via pro-inflammatory eicosanoid metabolism

    Vaittinen M, Lankinen MA, Käkelä P, et al. The FADS1 genotypes modify the effect of linoleic acid-enriched diet on adipose tissue inflammation via pro-inflammatory eicosanoid metabolism. European Journal of Nutrition61(7):3707–3718, 2022. doi:10.1007/s00394-022-02922-y

    work page doi:10.1007/s00394-022-02922-y 2022

  32. [32]

    The genetics of violence: Implications for research and policy.Journal of Law and the Biosciences1(2):221–225, 2014

    Appelbaum PS. The genetics of violence: Implications for research and policy.Journal of Law and the Biosciences1(2):221–225, 2014. doi:10.1093/jlb/lsu008

    work page doi:10.1093/jlb/lsu008 2014

  33. [33]

    Association of the dopamine D4 receptor (DRD4) gene and approach-related personality traits.Biological Psychiatry63(2):197–206,

    Munafò MR, Yalcin B, Willis-Owen SA, Flint J. Association of the dopamine D4 receptor (DRD4) gene and approach-related personality traits.Biological Psychiatry63(2):197–206,

  34. [34]

    doi:10.1016/j.biopsych.2007.04.006

    work page doi:10.1016/j.biopsych.2007.04.006 2007

  35. [35]

    Exploring the role of the Ser9Gly (rs6280) dopamine D3 receptor polymorphism in nicotine reinforcement and cue-elicited craving.Scien- tific Reports10:4085, 2020

    Chukwueke CC, Bhatt DL, George JT, et al. Exploring the role of the Ser9Gly (rs6280) dopamine D3 receptor polymorphism in nicotine reinforcement and cue-elicited craving.Scien- tific Reports10:4085, 2020. doi:10.1038/s41598-020-60940-4

    work page doi:10.1038/s41598-020-60940-4 2020

  36. [36]

    doi:10.1093/ije/dyv080

    BowdenJ,DaveySmithG,BurgessS.Mendelianrandomizationwithinvalidinstruments: effect estimation and bias detection through Egger regression.International Journal of Epidemiology 44(2):512–525, 2015. doi:10.1093/ije/dyv080

    work page doi:10.1093/ije/dyv080 2015

  37. [37]

    Inferring causal relationships between risk factors and outcomes from genome-wide association study data.Annual Review of Genomics and Human Genetics 19:303–327, 2018

    Burgess S, Foley CN, Zuber V. Inferring causal relationships between risk factors and outcomes from genome-wide association study data.Annual Review of Genomics and Human Genetics 19:303–327, 2018. doi:10.1146/annurev-genom-083117-021731. Declarations Conflict of Interest.Aruna Dey and Suraj Biswas are affiliated with Dots-In, an early-stage venture devel...

    work page doi:10.1146/annurev-genom-083117-021731 2018