{"id":703,"date":"2026-11-10T00:42:51","date_gmt":"2026-11-10T00:42:51","guid":{"rendered":"https:\/\/mrivas.su.domains\/gbe\/?p=703"},"modified":"2026-11-10T02:20:17","modified_gmt":"2026-11-10T02:20:17","slug":"the-elusive-search-for-genes-and-variants-that-increase-fluid-intelligence","status":"publish","type":"post","link":"https:\/\/mrivas.su.domains\/gbe\/uncategorized\/the-elusive-search-for-genes-and-variants-that-increase-fluid-intelligence\/","title":{"rendered":"The elusive search for genes and variants that increase fluid intelligence"},"content":{"rendered":"\n<style>.rivas-fluid-post{color:#231F20;background:#fff;padding:clamp(18px,4vw,48px);border-radius:12px;box-sizing:border-box}.rivas-fluid-post p,.rivas-fluid-post h2,.rivas-fluid-post h3,.rivas-fluid-post ol,.rivas-fluid-post .meta,.rivas-fluid-post .eyebrow{max-width:78ch;margin-left:auto;margin-right:auto}.rivas-fluid-post p{line-height:1.75}.rivas-fluid-post .dek{font-size:1.15em}.rivas-fluid-post .status{border-left:4px solid #913F8D;background:#f7f3f7;padding:1em}.rivas-fluid-post img{max-width:100%;height:auto}.rivas-fluid-post figure{margin:2em 0}.rivas-fluid-post figcaption{font-size:.85em;line-height:1.55;color:#4b4650}.rivas-fluid-post .tablewrap,.rivas-fluid-post .rivas-equation{overflow-x:auto}.rivas-fluid-post table{border-collapse:collapse;font-size:.85em;width:100%;margin:1.5em 0}.rivas-fluid-post th,.rivas-fluid-post td{border-bottom:1px solid #ECEDEE;text-align:left;vertical-align:top;padding:.65em}.rivas-fluid-post th{background:#ECEDEE}.rivas-fluid-post .rivas-equation{padding:.8em 0}.rivas-fluid-post .ql-img-displayed-equation{max-width:100%;height:auto;display:block;margin:1em auto}.rivas-fluid-post a{color:#135b77}.rivas-fluid-post .small,.rivas-fluid-post .meta{font-size:.85em}<\/style>\n<article class=\"rivas-fluid-post alignwide\">\n\n\n<p class=\"dek\">Positive genetic associations offer clues, but the route from a higher-score estimate to a convincing biological mechanism is difficult. MYEF2 shows both the promise and the uncertainty.<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/lrwpvki8f79lez0s1549s\/annotation_metaregression_concept.png?rlkey=d7ycwmyxafdqeuh64gb5gc78m&amp;raw=1\" target=\"_blank\" rel=\"noopener\"><img src=\"https:\/\/www.dropbox.com\/scl\/fi\/lrwpvki8f79lez0s1549s\/annotation_metaregression_concept.png?rlkey=d7ycwmyxafdqeuh64gb5gc78m&amp;raw=1\" alt=\"Conceptual illustration of integrating variant annotations\" width=\"1672\" height=\"941\" decoding=\"async\" loading=\"eager\" fetchpriority=\"high\"><\/a><figcaption class=\"wp-element-caption\">Conceptual illustration of integrating constraint, missense pathogenicity and loss-of-function annotations. The ribbons are artistic motifs, not measured protein structures or statistical results.<\/figcaption><\/figure>\n<p>Can human genetic variation identify biological changes associated with higher fluid-intelligence scores? Rare coding variants offer a way to investigate the question: they alter proteins through specific amino-acid substitutions, splice changes and truncations. Their effects can point in different directions within the same gene. A positive estimate becomes more persuasive when it survives scrutiny of consequence class, constraint, allele frequency, uncertainty and replication.<\/p>\n<p>We analyzed Genebass summary statistics for fluid intelligence in <strong>128,302 UK Biobank participants<\/strong>. The phenotype is the inverse-rank-normalized fluid-intelligence score, UK Biobank field 20016. We asked a focused question: within each gene, do variant effects differ systematically with HMM constraint, AlphaMissense pathogenicity, and high-confidence loss-of-function status?<\/p>\n<p>The frozen analysis contains <strong>4,977,720<\/strong> variant\u2013gene records from <strong>19,287<\/strong> retrieved genes; 19,215 genes support a testable joint baseline model. The archived coverage table identifies every unavailable target. In the primary analysis, <strong>2<\/strong> genes cross the planned threshold of <em>p<\/em> &lt; 2.58 \u00d7 10<sup>-6<\/sup>, corresponding to 0.05 divided by all 19,373 target genes. The genes crossing this threshold are <span class=\"gene\">ANKRD12<\/span>, <span class=\"gene\">KDM5B<\/span>.<\/p>\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/9www167hqbennrz76r8oe\/Fluid_Intelligence_Blog_Scan.png?rlkey=e6rozr4tp5nnj2z2dwmxw17q5&amp;raw=1\" target=\"_blank\" rel=\"noopener\"><img src=\"https:\/\/www.dropbox.com\/scl\/fi\/9www167hqbennrz76r8oe\/Fluid_Intelligence_Blog_Scan.png?rlkey=e6rozr4tp5nnj2z2dwmxw17q5&amp;raw=1\" alt=\"Joint annotation-model scan of fluid intelligence across 19,287 retrieved genes\" width=\"3564\" height=\"1608\" decoding=\"async\" loading=\"lazy\"><\/a><figcaption class=\"wp-element-caption\">Primary per-gene annotation F-test versus an intercept-only common effect, AF\u22640.05. The dashed line is 0.05\/19,373; all planned genes remain in this denominator. Points are genes. FWER primary hits are purple diamonds; 5% BH-FDR secondary hits are lime outlined diamonds; nominal MYEF2 is an orange star. BH FDR is computed within the 19,215 testable baseline joint-model genes. BH FDR uses only the baseline joint-model testable-gene family. Available-family BH classifications are conditional on the available source coverage. The QQ plot does not account for unavailable LD or carrier covariance.<\/figcaption><\/figure>\n\n<h2>A model for separating direction from annotation patterns<\/h2>\n<p>The analysis adapts the unified meta-regression framework developed by <a href=\"#ref-lauer\">Lauer and Rivas<\/a> and applied to rare-variant studies by <a href=\"#ref-aguilar\">Aguilar, Rivas, and Rivas<\/a>. The present specification uses three moderators and explicit annotation imputation; it is an adaptation rather than a literal replication of either preprint\u2019s full model. Each observation is a single-variant association estimate, not an individual participant. Within gene <em>g<\/em>, we fit:<\/p>\n<div class=\"rivas-equation\"><p class=\"ql-center-displayed-equation\" style=\"line-height: 56px;\"><span class=\"ql-right-eqno\"> &nbsp; <\/span><span class=\"ql-left-eqno\"> &nbsp; <\/span><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-82858fae75195ae728ad4cddadbf7116_l3.png\" height=\"56\" width=\"351\" class=\"ql-img-displayed-equation quicklatex-auto-format\" alt=\"&#92;&#91;&#92;&#98;&#101;&#103;&#105;&#110;&#123;&#97;&#108;&#105;&#103;&#110;&#101;&#100;&#125;&#92;&#119;&#105;&#100;&#101;&#104;&#97;&#116;&#92;&#98;&#101;&#116;&#97;&#95;&#123;&#105;&#103;&#125;&#38;&#61;&#92;&#97;&#108;&#112;&#104;&#97;&#95;&#103;&#43;&#92;&#103;&#97;&#109;&#109;&#97;&#95;&#123;&#72;&#44;&#103;&#125;&#72;&#95;&#123;&#105;&#103;&#125;&#43;&#92;&#103;&#97;&#109;&#109;&#97;&#95;&#123;&#65;&#44;&#103;&#125;&#65;&#95;&#123;&#105;&#103;&#125;&#43;&#92;&#103;&#97;&#109;&#109;&#97;&#95;&#123;&#76;&#44;&#103;&#125;&#76;&#95;&#123;&#105;&#103;&#125;&#43;&#92;&#118;&#97;&#114;&#101;&#112;&#115;&#105;&#108;&#111;&#110;&#95;&#123;&#105;&#103;&#125;&#44;&#92;&#92;&#32;&#119;&#95;&#123;&#105;&#103;&#125;&#38;&#61;&#92;&#119;&#105;&#100;&#101;&#104;&#97;&#116;&#32;&#115;&#95;&#123;&#105;&#103;&#125;&#94;&#123;&#45;&#50;&#125;&#44;&#92;&#113;&#113;&#117;&#97;&#100;&#92;&#111;&#112;&#101;&#114;&#97;&#116;&#111;&#114;&#110;&#97;&#109;&#101;&#123;&#86;&#97;&#114;&#125;&#40;&#92;&#118;&#97;&#114;&#101;&#112;&#115;&#105;&#108;&#111;&#110;&#95;&#123;&#105;&#103;&#125;&#41;&#61;&#92;&#115;&#105;&#103;&#109;&#97;&#95;&#103;&#94;&#50;&#92;&#119;&#105;&#100;&#101;&#104;&#97;&#116;&#32;&#115;&#95;&#123;&#105;&#103;&#125;&#94;&#123;&#92;&#44;&#50;&#125;&#46;&#92;&#101;&#110;&#100;&#123;&#97;&#108;&#105;&#103;&#110;&#101;&#100;&#125;&#92;&#93;\" title=\"Rendered by QuickLaTeX.com\"\/><\/p><\/div>\n<div class=\"tablewrap\"><table><thead><tr><th>Notation<\/th><th>Meaning<\/th><\/tr><\/thead><tbody><tr><td><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-8d96dadd220abde3c7c49a4d56d5fed1_l3.png\" class=\"ql-img-inline-formula quicklatex-auto-format\" alt=\"&#105;&#44;&#92;&#32;&#103;\" title=\"Rendered by QuickLaTeX.com\" height=\"16\" width=\"29\" style=\"vertical-align: -4px;\"\/><\/td><td>Variant index and gene index. The observations are variant association estimates.<\/td><\/tr><tr><td><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-c1fb134a53f86382ec340d2cfc8238a4_l3.png\" class=\"ql-img-inline-formula quicklatex-auto-format\" alt=\"&#92;&#119;&#105;&#100;&#101;&#104;&#97;&#116;&#92;&#98;&#101;&#116;&#97;&#95;&#123;&#105;&#103;&#125;&#44;&#92;&#32;&#112;&#95;&#123;&#105;&#103;&#125;&#44;&#92;&#32;&#92;&#119;&#105;&#100;&#101;&#104;&#97;&#116;&#32;&#115;&#95;&#123;&#105;&#103;&#125;\" title=\"Rendered by QuickLaTeX.com\" height=\"25\" width=\"92\" style=\"vertical-align: -6px;\"\/><\/td><td>Genebass variant effect, reported two-sided association p-value, and reconstructed standard error for the inverse-rank-normalized score.<\/td><\/tr><tr><td><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-a5abcb76b63cb29db3c76e3f084c5248_l3.png\" class=\"ql-img-inline-formula quicklatex-auto-format\" alt=\"&#72;&#95;&#123;&#105;&#103;&#125;&#44;&#92;&#32;&#65;&#95;&#123;&#105;&#103;&#125;&#44;&#92;&#32;&#76;&#95;&#123;&#105;&#103;&#125;\" title=\"Rendered by QuickLaTeX.com\" height=\"19\" width=\"105\" style=\"vertical-align: -6px;\"\/><\/td><td>Transformed HMM constraint and AlphaMissense scores, and observed HC-LoF indicator (or its explicitly imputed proportion).<\/td><\/tr><tr><td><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-ae71063e47cb5d799dc58032db2fb645_l3.png\" class=\"ql-img-inline-formula quicklatex-auto-format\" alt=\"&#92;&#97;&#108;&#112;&#104;&#97;&#95;&#103;&#44;&#92;&#32;&#92;&#103;&#97;&#109;&#109;&#97;&#95;&#123;&#72;&#44;&#103;&#125;&#44;&#92;&#32;&#92;&#103;&#97;&#109;&#109;&#97;&#95;&#123;&#65;&#44;&#103;&#125;&#44;&#92;&#32;&#92;&#103;&#97;&#109;&#109;&#97;&#95;&#123;&#76;&#44;&#103;&#125;\" title=\"Rendered by QuickLaTeX.com\" height=\"14\" width=\"153\" style=\"vertical-align: -6px;\"\/><\/td><td>Gene intercept and adjusted annotation slopes. The intercept refers to zero-valued predictors; it is not a missense-class coefficient.<\/td><\/tr><tr><td><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-f8f400bcf2fba8481af9acae7e693277_l3.png\" class=\"ql-img-inline-formula quicklatex-auto-format\" alt=\"&#92;&#118;&#97;&#114;&#101;&#112;&#115;&#105;&#108;&#111;&#110;&#95;&#123;&#105;&#103;&#125;&#44;&#92;&#32;&#92;&#115;&#105;&#103;&#109;&#97;&#95;&#103;&#94;&#50;&#44;&#92;&#32;&#119;&#95;&#123;&#105;&#103;&#125;\" title=\"Rendered by QuickLaTeX.com\" height=\"22\" width=\"92\" style=\"vertical-align: -7px;\"\/><\/td><td>Residual error, gene-specific residual variance scale, and inverse-variance weight.<\/td><\/tr><tr><td><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-75c787f184779b8707a0b0fe472c6b74_l3.png\" class=\"ql-img-inline-formula quicklatex-auto-format\" alt=\"&#110;&#95;&#103;&#44;&#92;&#32;&#114;&#95;&#103;\" title=\"Rendered by QuickLaTeX.com\" height=\"14\" width=\"47\" style=\"vertical-align: -6px;\"\/><\/td><td>Usable variant count and rank of the full design matrix. Residual degrees of freedom are n_g minus r_g.<\/td><\/tr><tr><td><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-6b87e8f1e93ceb7902f74f7cc4386623_l3.png\" class=\"ql-img-inline-formula quicklatex-auto-format\" alt=\"&#92;&#109;&#97;&#116;&#104;&#114;&#109;&#123;&#82;&#83;&#83;&#125;&#95;&#123;&#48;&#103;&#125;&#44;&#92;&#32;&#92;&#109;&#97;&#116;&#104;&#114;&#109;&#123;&#82;&#83;&#83;&#125;&#95;&#123;&#49;&#103;&#125;&#44;&#92;&#32;&#70;&#95;&#103;\" title=\"Rendered by QuickLaTeX.com\" height=\"18\" width=\"140\" style=\"vertical-align: -6px;\"\/><\/td><td>Weighted residual sums of squares for the intercept-only and annotation models, and their rank-aware F statistic.<\/td><\/tr><tr><td><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-80dbde0850a441877e4e81c0e74b6fa8_l3.png\" class=\"ql-img-inline-formula quicklatex-auto-format\" alt=\"&#92;&#80;&#104;&#105;&#94;&#123;&#45;&#49;&#125;&#44;&#92;&#32;&#92;&#109;&#97;&#116;&#104;&#114;&#109;&#123;&#65;&#70;&#125;\" title=\"Rendered by QuickLaTeX.com\" height=\"19\" width=\"69\" style=\"vertical-align: -4px;\"\/><\/td><td>Standard-normal quantile function and source phenotype-specific allele frequency. AF cutoffs select nested sensitivity sets.<\/td><\/tr><\/tbody><\/table><\/div>\n<p>HMM summarizes regional constraint; AlphaMissense supplies a predicted missense pathogenicity score; and the loss-of-function term distinguishes variants labeled high confidence by LOFTEE. We clip HMM and AlphaMissense probabilities to [0.01, 0.99] and transform them as \u2212log(1 \u2212 score). The fit gives more weight to effects estimated with greater precision.<\/p>\n<p>The primary F-test compares this model with an intercept-only model. Its null is <img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-e85e79de3020146626e789e4caa9432b_l3.png\" class=\"ql-img-inline-formula quicklatex-auto-format\" alt=\"&#72;&#95;&#123;&#48;&#44;&#103;&#125;&#58;&#92;&#32;&#92;&#103;&#97;&#109;&#109;&#97;&#95;&#123;&#72;&#44;&#103;&#125;&#61;&#92;&#103;&#97;&#109;&#109;&#97;&#95;&#123;&#65;&#44;&#103;&#125;&#61;&#92;&#103;&#97;&#109;&#109;&#97;&#95;&#123;&#76;&#44;&#103;&#125;&#61;&#48;\" title=\"Rendered by QuickLaTeX.com\" height=\"18\" width=\"228\" style=\"vertical-align: -6px;\"\/>, for the annotation contrasts that can be estimated. A small <em>p<\/em>-value means that annotations help explain differences among variant effects. The fitted coefficients and observed effect signs supply the directional evidence.<\/p>\n<div class=\"rivas-equation\"><p class=\"ql-center-displayed-equation\" style=\"line-height: 44px;\"><span class=\"ql-right-eqno\"> &nbsp; <\/span><span class=\"ql-left-eqno\"> &nbsp; <\/span><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-ad5c873630dfaef9e830edd95ad83aab_l3.png\" height=\"44\" width=\"379\" class=\"ql-img-displayed-equation quicklatex-auto-format\" alt=\"&#92;&#91;&#70;&#95;&#103;&#61;&#92;&#102;&#114;&#97;&#99;&#123;&#40;&#92;&#109;&#97;&#116;&#104;&#114;&#109;&#123;&#82;&#83;&#83;&#125;&#95;&#123;&#48;&#103;&#125;&#45;&#92;&#109;&#97;&#116;&#104;&#114;&#109;&#123;&#82;&#83;&#83;&#125;&#95;&#123;&#49;&#103;&#125;&#41;&#47;&#40;&#114;&#95;&#103;&#45;&#49;&#41;&#125;&#123;&#92;&#109;&#97;&#116;&#104;&#114;&#109;&#123;&#82;&#83;&#83;&#125;&#95;&#123;&#49;&#103;&#125;&#47;&#40;&#110;&#95;&#103;&#45;&#114;&#95;&#103;&#41;&#125;&#92;&#32;&#92;&#111;&#118;&#101;&#114;&#115;&#101;&#116;&#123;&#72;&#95;&#48;&#125;&#123;&#92;&#115;&#105;&#109;&#125;&#92;&#32;&#70;&#95;&#123;&#114;&#95;&#103;&#45;&#49;&#44;&#92;&#44;&#110;&#95;&#103;&#45;&#114;&#95;&#103;&#125;&#46;&#92;&#93;\" title=\"Rendered by QuickLaTeX.com\"\/><\/p><\/div>\n<p>We estimate the residual scale and use analytic, rank-aware F- and t-tests. When an annotation is constant within a gene, its separate coefficient cannot be identified; the joint test uses the remaining independent contrasts. The detailed results include the three single-annotation models and confidence intervals for estimable coefficients.<\/p>\n<aside><strong>Why an annotation test and a burden test can disagree.<\/strong> A Genebass burden test asks whether a weighted aggregate of variants is associated with the phenotype. Our F-test asks whether annotation-related differences improve the description of their effects. A gene can have a burden association with little variation by annotation, or annotation-related variation with a modest aggregate effect. These are complementary questions.<\/aside>\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/a4a6g7a8tb1jin3jhzr5t\/Fluid_Intelligence_Blog_Model.png?rlkey=0xv7h0bm51rodbi6cujo09op0&amp;raw=1\" target=\"_blank\" rel=\"noopener\"><img src=\"https:\/\/www.dropbox.com\/scl\/fi\/a4a6g7a8tb1jin3jhzr5t\/Fluid_Intelligence_Blog_Model.png?rlkey=0xv7h0bm51rodbi6cujo09op0&amp;raw=1\" alt=\"The three-moderator meta-regression model and imputation procedure\" width=\"3331\" height=\"1980\" decoding=\"async\" loading=\"lazy\"><\/a><figcaption class=\"wp-element-caption\">The actual three-moderator model used in this scan, adapted from Lauer and Rivas and Aguilar, Rivas and Rivas. SE recovery is an approximate normal inversion of beta and the reported two-sided p-value. Missing annotations are mean-imputed; source effects are never imputed. The rank-aware residual-scaled F-test compares the full annotation model to an intercept-only model, not a zero-effect burden null.<\/figcaption><\/figure>\n\n<h2>The clearer evidence points toward lower scores<\/h2>\n<p><span class=\"gene\">ANKRD12<\/span>, already implicated in adult cognitive function by <a href=\"https:\/\/doi.org\/10.1038\/s41588-023-01398-8\">Chen et al. (2023)<\/a>, has a joint-model result of <strong>F(3, 886) = 17.45, <em>p<\/em> = 5.14 \u00d7 10<sup>-11<\/sup><\/strong> across 890 baseline variants. It meets the primary planned-familywise criterion. The source Genebass HC LoF burden effect is negative, indicating association with lower scores (<em>p<\/em> = 5.28 \u00d7 10<sup>-14<\/sup>). Source-matched annotations identify 53 observed HC LoF variants, with 13 positive and 40 negative estimates; 0 LoF values remain imputed. The varying observed LoF indicator supplies information that constant whole-gene imputation cannot create. This is an annotation-based reanalysis of UK Biobank data, with cohort overlap rather than independent replication.<\/p>\n<p><span class=\"gene\">KDM5B<\/span> encodes an H3K4 histone demethylase, an enzyme that removes methyl marks from histone H3 (<a href=\"https:\/\/doi.org\/10.1016\/j.molcel.2007.03.001\">Yamane et al., 2007<\/a>). Its relationship to adult cognition is already established: <a href=\"https:\/\/doi.org\/10.1038\/s41588-023-01398-8\">Chen et al. (2023)<\/a> associated rare protein-truncating variants with lower cognitive performance, including verbal-numerical reasoning, in UK Biobank. The present work revisits UK Biobank summary statistics through an annotation model. It adds a view of differences among variant effects; it does not establish a new cognition gene or provide replication in an independent cohort.<\/p>\n<p>For <span class=\"gene\">KDM5B<\/span>, the joint model gives <strong>F(3, 626) = 10.81, <em>p<\/em> = 6.24 \u00d7 10<sup>-7<\/sup><\/strong>; the Bonferroni-adjusted value across all planned genes is 0.01208. This crosses the planned exome-wide primary threshold. Its source Genebass high-confidence loss-of-function burden effect is <strong>negative<\/strong> (<em>p<\/em> = 1.38 \u00d7 10<sup>-6<\/sup>), indicating an association with lower fluid-intelligence scores.<\/p>\n<p>The adjusted loss-of-function contrast in the <span class=\"gene\">KDM5B<\/span> meta-regression is -0.538 (95% CI -0.752 to -0.323). This coefficient compares high-confidence loss-of-function variants with the model\u2019s non-LoF reference at the same HMM and AlphaMissense values. It is a different quantity from the published burden effect, so the two magnitudes should not be interpreted interchangeably.<\/p>\n<h2>MYEF2: following the positive missense clue<\/h2>\n<p><span class=\"gene\">MYEF2<\/span> illustrates why both direction and the statistical question matter. Its published Genebass loss-of-function burden association is negative (<em>p<\/em> = 0.001207), while its missense\/low-confidence-LoF burden association is positive (<em>p<\/em> = 0.003002). Both source burden <em>p<\/em>-values are below 0.01. The annotation model, using 220 variants, gives F(3, 216) = 8.032 and <em>p<\/em> = 4.25 \u00d7 10<sup>-5<\/sup>. This does not cross the planned exome-wide primary threshold or the secondary BH-FDR criterion; we show it as a nominal association and an illustrative case study.<\/p>\n<p>To examine the positive missense clue directly, we fit a focused consequence-class meta-regression in <span class=\"gene\">MYEF2<\/span>, keeping all 220 baseline variants and their QC flags. The <strong>159 strict missense variants have a positive class coefficient of 0.0662<\/strong> (95% CI 0.0162 to 0.116; nominal <em>p<\/em> = 0.00966). The observed HC-LoF class coefficient is -0.0957. Their missense-minus-LoF contrast is <strong>0.162<\/strong> (95% CI 0.0834 to 0.24; <em>p<\/em> = 6.67e-05).<\/p><p>This is a secondary analysis prompted by the MYEF2 case study. Each class coefficient is a precision-weighted mean of single-variant effects. The model has one coefficient per consequence class and no additional intercept. Its class-mean tests concern zero, while the exome-wide annotation F-test concerns differences explained by annotations.<\/p><div class=\"rivas-equation\"><p class=\"ql-center-displayed-equation\" style=\"line-height: 120px;\"><span class=\"ql-right-eqno\"> &nbsp; <\/span><span class=\"ql-left-eqno\"> &nbsp; <\/span><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-072ce9033960990e23a2a922eec18d29_l3.png\" height=\"120\" width=\"527\" class=\"ql-img-displayed-equation quicklatex-auto-format\" alt=\"&#92;&#91;&#92;&#98;&#101;&#103;&#105;&#110;&#123;&#97;&#108;&#105;&#103;&#110;&#101;&#100;&#125;&#92;&#119;&#105;&#100;&#101;&#104;&#97;&#116;&#92;&#98;&#101;&#116;&#97;&#95;&#123;&#105;&#103;&#125;&#38;&#61;&#92;&#115;&#117;&#109;&#95;&#123;&#99;&#92;&#105;&#110;&#92;&#109;&#97;&#116;&#104;&#99;&#97;&#108;&#32;&#67;&#125;&#92;&#116;&#104;&#101;&#116;&#97;&#95;&#123;&#99;&#44;&#103;&#125;&#92;&#44;&#92;&#109;&#97;&#116;&#104;&#98;&#102;&#32;&#49;&#40;&#67;&#95;&#123;&#105;&#103;&#125;&#61;&#99;&#41;&#43;&#92;&#118;&#97;&#114;&#101;&#112;&#115;&#105;&#108;&#111;&#110;&#95;&#123;&#105;&#103;&#125;&#44;&#92;&#92;&#32;&#92;&#119;&#105;&#100;&#101;&#104;&#97;&#116;&#92;&#98;&#101;&#116;&#97;&#95;&#123;&#105;&#103;&#125;&#38;&#61;&#92;&#115;&#117;&#109;&#95;&#123;&#99;&#92;&#105;&#110;&#92;&#109;&#97;&#116;&#104;&#99;&#97;&#108;&#32;&#67;&#125;&#92;&#116;&#104;&#101;&#116;&#97;&#94;&#123;&#92;&#109;&#97;&#116;&#104;&#114;&#109;&#123;&#97;&#100;&#106;&#125;&#125;&#95;&#123;&#99;&#44;&#103;&#125;&#92;&#44;&#92;&#109;&#97;&#116;&#104;&#98;&#102;&#32;&#49;&#40;&#67;&#95;&#123;&#105;&#103;&#125;&#61;&#99;&#41;&#43;&#92;&#101;&#116;&#97;&#95;&#123;&#72;&#44;&#103;&#125;&#40;&#72;&#95;&#123;&#105;&#103;&#125;&#45;&#92;&#98;&#97;&#114;&#32;&#72;&#95;&#103;&#41;&#43;&#92;&#101;&#116;&#97;&#95;&#123;&#65;&#44;&#103;&#125;&#40;&#65;&#95;&#123;&#105;&#103;&#125;&#45;&#92;&#98;&#97;&#114;&#32;&#65;&#95;&#103;&#41;&#43;&#92;&#118;&#97;&#114;&#101;&#112;&#115;&#105;&#108;&#111;&#110;&#95;&#123;&#105;&#103;&#125;&#44;&#92;&#92;&#32;&#92;&#68;&#101;&#108;&#116;&#97;&#95;&#123;&#77;&#44;&#76;&#125;&#38;&#61;&#92;&#116;&#104;&#101;&#116;&#97;&#95;&#123;&#77;&#44;&#103;&#125;&#45;&#92;&#116;&#104;&#101;&#116;&#97;&#95;&#123;&#76;&#44;&#103;&#125;&#46;&#92;&#101;&#110;&#100;&#123;&#97;&#108;&#105;&#103;&#110;&#101;&#100;&#125;&#92;&#93;\" title=\"Rendered by QuickLaTeX.com\"\/><\/p><\/div><p>Here, C<sub>ig<\/sub> is the verified consequence class, and the class set contains strict missense (M), synonymous, observed HC LoF (L), and other coding consequences. The indicator is one for membership in a class and zero otherwise. The coefficients \u03b8 summarize class means; \u03b8<sup>adj<\/sup> describes means at common annotation reference values. The barred H and A are the frozen gene baseline averages, \u03b7 are the additional annotation slopes, and \u0394<sub>M,L<\/sub> is the missense-minus-HC-LoF contrast. Both models use the same reconstructed uncertainties and residual-scaled t intervals as the primary framework.<\/p><div class=\"tablewrap\"><table><thead><tr><th>Consequence class<\/th><th>Variants<\/th><th>Mean effect<\/th><th>95% CI<\/th><th>Nominal p<\/th><th>Holm p (12 tests)<\/th><\/tr><\/thead><tbody><tr><td>Strict missense<\/td><td>159<\/td><td>0.0662<\/td><td>0.0162 to 0.116<\/td><td>0.00966<\/td><td>0.058<\/td><\/tr><tr><td>Synonymous<\/td><td>43<\/td><td>0.0454<\/td><td>-0.0704 to 0.161<\/td><td>0.44<\/td><td>1<\/td><\/tr><tr><td>Observed HC LoF<\/td><td>12<\/td><td>-0.0957<\/td><td>-0.156 to -0.0352<\/td><td>0.00206<\/td><td>0.0207<\/td><\/tr><tr><td>Other coding consequences<\/td><td>6<\/td><td>0.777<\/td><td>0.214 to 1.34<\/td><td>0.00709<\/td><td>0.0496<\/td><\/tr><\/tbody><\/table><\/div><p>The increasing-effect interpretation remains tentative. The missense coefficient has Holm-adjusted <em>p<\/em> = <strong>0.058<\/strong> across all 12 reported baseline class-mean and contrast tests. At common HMM and AlphaMissense values, its adjusted mean is 0.0489 (95% CI -0.0215 to 0.119; <em>p<\/em> = 0.172). The adjusted missense-minus-LoF contrast remains positive (0.144; <em>p<\/em> = 0.00232), while the adjusted missense-minus-synonymous contrast is -0.00265 (<em>p<\/em> = 0.969). Thus the positive mean gives limited evidence for a missense-specific increasing effect relative to synonymous variation.<\/p><p>The AlphaMissense moderator describes a separate severity gradient. Its single-annotation slope is -0.0824 (<em>p<\/em> = 0.00816), and its slope in the joint HMM\u2013AlphaMissense\u2013LoF model is -0.0291 (<em>p<\/em> = 0.388). These negative slopes and their adjustment dependence must accompany the positive missense-class mean. The published positive missense\/LC burden uses a broader consequence mask and a different aggregate estimand. None of these coefficients estimates the cognitive effect of editing a variant in a person.<\/p>\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/kwjndaojrgehanuzt14fv\/Fluid_Intelligence_Blog_MYEF2_Focus.png?rlkey=3sf1bi7w0p3kh9mqdfpqsr5m6&amp;raw=1\" target=\"_blank\" rel=\"noopener\"><img src=\"https:\/\/www.dropbox.com\/scl\/fi\/kwjndaojrgehanuzt14fv\/Fluid_Intelligence_Blog_MYEF2_Focus.png?rlkey=3sf1bi7w0p3kh9mqdfpqsr5m6&amp;raw=1\" alt=\"MYEF2 consequence-class means, contrasts, severity slopes, and allele-frequency sensitivities\" width=\"3447\" height=\"2562\" decoding=\"async\" loading=\"lazy\"><\/a><figcaption class=\"wp-element-caption\">MYEF2 focused secondary consequence-class moderator analysis. A: all four class means versus zero. B: missense-minus-LoF and missense-minus-synonymous contrasts, before and after adjustment for centered HMM\/AlphaMissense. C: the original AlphaMissense-only and joint severity slopes. D: missense and HC-LoF class means across nested AF subsets. Nominal p-values and residual-scaled 95% t intervals are shown; 12-test within-AF Holm values accompany the tables. All source variants and QC flags remain included. Positive means do not establish cognitive enhancement.<\/figcaption><\/figure>\n<p>Rodent studies connect Myef2 to myelin-gene regulation through DNA-binding repression of <em>Mbp<\/em> and RNA interactions with <em>Mbp<\/em> and <em>Mag<\/em> in oligodendrocytes (<a href=\"https:\/\/doi.org\/10.1074\/jbc.270.21.12503\">Haas et al., 1995<\/a>; <a href=\"https:\/\/doi.org\/10.1016\/j.celrep.2021.109695\">Wang et al., 2021<\/a>). Whether these mechanisms explain the human association remains unestablished.<\/p>\n<p>Individual loss-of-function estimates need not all share the aggregate sign. Among the observed high-confidence loss-of-function variants, <span class=\"gene\">KDM5B<\/span> has 14 positive and 40 negative estimates; <span class=\"gene\">MYEF2<\/span> has 7 positive and 5 negative estimates. Precision, allele frequency, and weighting matter. We therefore show individual effects alongside the source burden direction rather than assigning one sign to every variant in a gene.<\/p>\n<p>The descriptive high-confidence-LoF average for <span class=\"gene\">MYEF2<\/span> is especially concentrated: one negative splice-donor variant, <code>15-48152233-C-A<\/code>, contributes <strong>98.37% of the inverse-variance information<\/strong> among the 12 observed HC variants. It carries a retained interval-QC flag and imputed HMM and AlphaMissense values, and has no defensible exact residue placement. Its AF is 0.00472, so it is absent from both rarer subsets. This fraction describes our single-variant IVW summary; it does not apportion the published Genebass burden test or the joint F-test.<\/p>\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/xuqnrteyxzz2sgkly2oji\/Fluid_Intelligence_Blog_Direction.png?rlkey=uii5hr2ypzi9fgsw0crapw6a5&amp;raw=1\" target=\"_blank\" rel=\"noopener\"><img src=\"https:\/\/www.dropbox.com\/scl\/fi\/xuqnrteyxzz2sgkly2oji\/Fluid_Intelligence_Blog_Direction.png?rlkey=uii5hr2ypzi9fgsw0crapw6a5&amp;raw=1\" alt=\"Loss-of-function effect directions, source burden tests, and adjusted meta-regression coefficients\" width=\"3475\" height=\"1887\" decoding=\"async\" loading=\"lazy\"><\/a><figcaption class=\"wp-element-caption\">A: published Genebass pLoF burden beta and p-value, with no reconstructed confidence interval. B: HC LoF coefficient adjusted for HMM and AlphaMissense, with the model residual-scaled 95% t interval. A and B are distinct scales, variant masks and null hypotheses. C: signs of observed individual HC LoF variant effects; counts do not reflect precision. Known annotations only are used for counts; imputed fractional LoF values do not become known variants. Rows separate FWER primary, 5% baseline joint-model BH-FDR secondary, and nominal illustrative genes. BH FDR uses only the baseline joint-model testable-gene family. Available-family BH classifications are conditional on the available source coverage.<\/figcaption><\/figure>\n<div class=\"tablewrap\"><table><thead><tr><th>Gene<\/th><th>Variants<\/th><th>Joint F-test p<\/th><th>Planned Bonferroni p<\/th><th>Source LoF burden direction<\/th><th>Interpretation<\/th><\/tr><\/thead><tbody><tr><td><span class=\"gene\">ANKRD12<\/span><\/td><td>890<\/td><td>5.14 \u00d7 10<sup>-11<\/sup><\/td><td>9.96 \u00d7 10<sup>-7<\/sup><\/td><td>Lower score<\/td><td>Primary threshold passed<\/td><\/tr><tr><td><span class=\"gene\">KDM5B<\/span><\/td><td>630<\/td><td>6.24 \u00d7 10<sup>-7<\/sup><\/td><td>0.0121<\/td><td>Lower score<\/td><td>Primary threshold passed<\/td><\/tr><tr><td><span class=\"gene\">MYEF2<\/span><\/td><td>220<\/td><td>4.25 \u00d7 10<sup>-5<\/sup><\/td><td>0.822<\/td><td>Lower score<\/td><td>Nominal \/ illustrative<\/td><\/tr><\/tbody><\/table><\/div><p class=\"table-note\">Baseline AF \u2264 0.05. Correction covers the planned primary family of 19,373 genes. Individual moderator and sensitivity tests are not additional independent discoveries.<\/p>\n\n<h2>A secondary view at 5% false discovery rate<\/h2>\n<p>Benjamini\u2013Hochberg q-values use the 19,215 retrieved testable genes in the joint-model, AF \u2264 0.05 family. These secondary classifications are conditional on the available source data; 86 target genes remain untested. The primary Bonferroni denominator retains all planned genes. At <strong>q \u2264 0.05<\/strong>, 3 genes meet the secondary criterion; 2 of these also pass the primary planned-familywise threshold, while 1 is an additional FDR candidate. The table below includes the full secondary set and labels the overlap explicitly.<\/p>\n<div class=\"tablewrap\"><table><thead><tr><th>Gene<\/th><th>Variants<\/th><th>Joint F-test p<\/th><th>BH q<\/th><th>Planned Bonferroni p<\/th><th>Source LoF burden direction<\/th><th>Evidence<\/th><\/tr><\/thead><tbody><tr><td><span class=\"gene\">ANKRD12<\/span><\/td><td>890<\/td><td>5.14 \u00d7 10<sup>-11<\/sup><\/td><td>9.88 \u00d7 10<sup>-7<\/sup><\/td><td>9.96 \u00d7 10<sup>-7<\/sup><\/td><td>Lower score<\/td><td>Primary + secondary<\/td><\/tr><tr><td><span class=\"gene\">KDM5B<\/span><\/td><td>630<\/td><td>6.24 \u00d7 10<sup>-7<\/sup><\/td><td>0.00599<\/td><td>0.0121<\/td><td>Lower score<\/td><td>Primary + secondary<\/td><\/tr><tr><td><span class=\"gene\">FBRSL1<\/span><\/td><td>922<\/td><td>3.91 \u00d7 10<sup>-6<\/sup><\/td><td>0.0251<\/td><td>0.0758<\/td><td>Higher score<\/td><td>Secondary FDR only<\/td><\/tr><\/tbody><\/table><\/div>\n<p><span class=\"gene\">FBRSL1<\/span> illustrates the secondary set. Its source Genebass LoF burden sign is positive (<em>p<\/em> = 0.0246), and its adjusted LoF contrast is 0.0369 (95% CI 0.019 to 0.0547). Among 31 observed HC LoF variants, 24 estimates are positive and 7 negative. It does not meet the primary planned-familywise criterion (Bonferroni <em>p<\/em> = 0.0758). The joint-test evidence weakens from <em>p<\/em> = 3.91 \u00d7 10<sup>-6<\/sup> at AF \u2264 0.05 to 0.0314 at AF &lt; 0.001 and 0.0263 at AF &lt; 0.0001. This candidate therefore depends substantially on the broader baseline variant set.<\/p>\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/uzjdeq5oiunkr7dqxrpe1\/Fluid_Intelligence_Blog_Moderators.png?rlkey=riwd45ychglj5vbcqrvn7em5o&amp;raw=1\" target=\"_blank\" rel=\"noopener\"><img src=\"https:\/\/www.dropbox.com\/scl\/fi\/uzjdeq5oiunkr7dqxrpe1\/Fluid_Intelligence_Blog_Moderators.png?rlkey=riwd45ychglj5vbcqrvn7em5o&amp;raw=1\" alt=\"Adjusted HMM, AlphaMissense, and loss-of-function moderator coefficients\" width=\"3469\" height=\"1887\" decoding=\"async\" loading=\"lazy\"><\/a><figcaption class=\"wp-element-caption\">Joint-model HMM, AlphaMissense and HC LoF coefficients with residual-scaled 95% t confidence intervals. All coefficients adjust for the other two moderators and an intercept. Predictor units and axis ranges differ; coefficient magnitudes are not importance scores. Rows are selected by baseline joint-model FWER or BH-FDR status plus requested illustrative anchors, never by single-moderator p-values. BH FDR uses only the baseline joint-model testable-gene family. Available-family BH classifications are conditional on the available source coverage.<\/figcaption><\/figure>\n<p>The primary Bonferroni criterion and the secondary BH-FDR criterion serve different purposes. The first is our stricter primary discovery threshold across all planned genes. The second identifies a broader set for exploration under the BH procedure and the model\u2019s calibration assumptions. A secondary FDR candidate is labeled as such throughout the figures and structural views. HMM-only, AlphaMissense-only, LoF-only, and rarer-AF analyses are separate test families and do not enter this secondary set.<\/p>\n\n<h2>Testing whether candidate genes converge on pathways<\/h2>\n<p>We selected all <strong>48 genes with raw joint-model <em>p<\/em> &lt; 0.001<\/strong> at AF \u2264 0.05, then tested Reactome release 98 human pathways. 30 selected genes map to Reactome. The background is the 11,259 testable genes represented in that same mapping, drawn from 19,215 testable genes overall. We tested 1,753 pathways containing 10\u2013500 background genes and included zero-overlap pathways in the multiple-testing family. The F-test is unsigned, so this gene set mixes higher- and lower-score associations.<\/p><div class=\"rivas-equation\"><p class=\"ql-center-displayed-equation\" style=\"line-height: 58px;\"><span class=\"ql-right-eqno\"> &nbsp; <\/span><span class=\"ql-left-eqno\"> &nbsp; <\/span><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-570a23fb5b626f3adb4f4797024dc494_l3.png\" height=\"58\" width=\"660\" class=\"ql-img-displayed-equation quicklatex-auto-format\" alt=\"&#92;&#91;&#88;&#95;&#80;&#92;&#115;&#105;&#109;&#92;&#111;&#112;&#101;&#114;&#97;&#116;&#111;&#114;&#110;&#97;&#109;&#101;&#123;&#72;&#121;&#112;&#101;&#114;&#103;&#101;&#111;&#109;&#125;&#40;&#78;&#95;&#66;&#44;&#75;&#95;&#80;&#44;&#78;&#95;&#70;&#41;&#44;&#92;&#113;&#113;&#117;&#97;&#100;&#32;&#112;&#95;&#80;&#61;&#92;&#80;&#114;&#40;&#88;&#95;&#80;&#92;&#103;&#101;&#32;&#120;&#95;&#80;&#41;&#61;&#92;&#115;&#117;&#109;&#95;&#123;&#106;&#61;&#120;&#95;&#80;&#125;&#94;&#123;&#92;&#109;&#105;&#110;&#40;&#75;&#95;&#80;&#44;&#78;&#95;&#70;&#41;&#125;&#92;&#102;&#114;&#97;&#99;&#123;&#92;&#98;&#105;&#110;&#111;&#109;&#123;&#75;&#95;&#80;&#125;&#123;&#106;&#125;&#92;&#98;&#105;&#110;&#111;&#109;&#123;&#78;&#95;&#66;&#45;&#75;&#95;&#80;&#125;&#123;&#78;&#95;&#70;&#45;&#106;&#125;&#125;&#123;&#92;&#98;&#105;&#110;&#111;&#109;&#123;&#78;&#95;&#66;&#125;&#123;&#78;&#95;&#70;&#125;&#125;&#46;&#92;&#93;\" title=\"Rendered by QuickLaTeX.com\"\/><\/p><\/div><p>For pathway P, N<sub>B<\/sub> is the mapped background size, N<sub>F<\/sub> the mapped selected-gene count, K<sub>P<\/sub> the pathway\u2019s background-gene count, and x<sub>P<\/sub> the observed selected-gene overlap. X<sub>P<\/sub> is the random overlap under sampling from that background; j indexes possible overlaps, and the binomial coefficients count combinations. The upper-tail probability p<sub>P<\/sub> tests over-representation. We apply Benjamini\u2013Hochberg correction across all eligible pathways to obtain q<sub>P<\/sub>. Fold enrichment is the observed selected-gene pathway fraction divided by the background fraction.<\/p><p><strong>No pathway meets BH q \u2264 0.05.<\/strong> The leading nominal pathway is <a href=\"https:\/\/reactome.org\/PathwayBrowser\/#\/R-HSA-112308\">Presynaptic depolarization and calcium channel opening<\/a>, with 2 selected genes among 11 background members: CACNA1A, CACNA1E. Its nominal <em>p<\/em> is 3.72e-04, and its BH q-value is 0.652. The corrected result determines the strength of this pathway evidence.<\/p><p>18 selected genes lack a Reactome mapping in this release: RC3H2, DNAAF6, PUM3, ANKRD12, MYEF2, PRDM13, FBRSL1, SLC35A5, TBX19, LNPK, MAP11, GIGYF1, CIART, CCDC112, NLRP7, EDDM3A, CGB7, RNF148. They remain in the foreground file and gene-level analysis. The enrichment uses the mapped subset. Pathway overlap, annotation coverage, gene size and unequal detection power further limit a thresholded enrichment analysis.<\/p><p>Several leading nominal pathways repeatedly contain the same genes. These shared overlaps generate related pathway labels and should be read as correlated annotations of the gene list.<\/p><div class=\"tablewrap\"><table><thead><tr><th>Reactome pathway<\/th><th>Overlap<\/th><th>Fold enrichment<\/th><th>Nominal p<\/th><th>BH q<\/th><th>Selected genes<\/th><\/tr><\/thead><tbody><tr><td><a href=\"https:\/\/reactome.org\/PathwayBrowser\/#\/R-HSA-112308\">Presynaptic depolarization and calcium channel opening<\/a><\/td><td>2\/11<\/td><td>68.2<\/td><td>3.72e-04<\/td><td>0.652<\/td><td>CACNA1A, CACNA1E<\/td><\/tr><tr><td><a href=\"https:\/\/reactome.org\/PathwayBrowser\/#\/R-HSA-168271\">Transport of Ribonucleoproteins into the Host Nucleus<\/a><\/td><td>2\/32<\/td><td>23.5<\/td><td>0.00324<\/td><td>1<\/td><td>KPNB1, NUP88<\/td><\/tr><tr><td><a href=\"https:\/\/reactome.org\/PathwayBrowser\/#\/R-HSA-180746\">Nuclear import of Rev protein<\/a><\/td><td>2\/34<\/td><td>22.1<\/td><td>0.00365<\/td><td>1<\/td><td>KPNB1, NUP88<\/td><\/tr><tr><td><a href=\"https:\/\/reactome.org\/PathwayBrowser\/#\/R-HSA-162909\">Host Interactions of HIV factors<\/a><\/td><td>3\/118<\/td><td>9.54<\/td><td>0.00371<\/td><td>1<\/td><td>PSMC5, KPNB1, NUP88<\/td><\/tr><tr><td><a href=\"https:\/\/reactome.org\/PathwayBrowser\/#\/R-HSA-177243\">Interactions of Rev with host cellular proteins<\/a><\/td><td>2\/37<\/td><td>20.3<\/td><td>0.00431<\/td><td>1<\/td><td>KPNB1, NUP88<\/td><\/tr><tr><td><a href=\"https:\/\/reactome.org\/PathwayBrowser\/#\/R-HSA-168276\">NS1 Mediated Effects on Host Pathways<\/a><\/td><td>2\/43<\/td><td>17.5<\/td><td>0.00579<\/td><td>1<\/td><td>KPNB1, NUP88<\/td><\/tr><tr><td><a href=\"https:\/\/reactome.org\/PathwayBrowser\/#\/R-HSA-202733\">Cell surface interactions at the vascular wall<\/a><\/td><td>3\/141<\/td><td>7.99<\/td><td>0.0061<\/td><td>1<\/td><td>SLC7A8, CXADR, CD47<\/td><\/tr><tr><td><a href=\"https:\/\/reactome.org\/PathwayBrowser\/#\/R-HSA-75153\">Apoptotic execution phase<\/a><\/td><td>2\/52<\/td><td>14.4<\/td><td>0.00838<\/td><td>1<\/td><td>KPNB1, PLEC<\/td><\/tr><tr><td><a href=\"https:\/\/reactome.org\/PathwayBrowser\/#\/R-HSA-109581\">Apoptosis<\/a><\/td><td>3\/165<\/td><td>6.82<\/td><td>0.00939<\/td><td>1<\/td><td>PSMC5, KPNB1, PLEC<\/td><\/tr><tr><td><a href=\"https:\/\/reactome.org\/PathwayBrowser\/#\/R-HSA-5619115\">Disorders of transmembrane transporters<\/a><\/td><td>3\/165<\/td><td>6.82<\/td><td>0.00939<\/td><td>1<\/td><td>SLC12A3, PSMC5, NUP88<\/td><\/tr><\/tbody><\/table><\/div>\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/911vq2f0644dczdava6kc\/Fluid_Intelligence_Blog_Pathways.png?rlkey=gaqdaigudi6i40gyhaokhv0tc&amp;raw=1\" target=\"_blank\" rel=\"noopener\"><img src=\"https:\/\/www.dropbox.com\/scl\/fi\/911vq2f0644dczdava6kc\/Fluid_Intelligence_Blog_Pathways.png?rlkey=gaqdaigudi6i40gyhaokhv0tc&amp;raw=1\" alt=\"Reactome enrichment of 48 genes with joint-model p below 0.001; no pathway passes FDR correction\" width=\"3331\" height=\"2074\" decoding=\"async\" loading=\"lazy\"><\/a><figcaption class=\"wp-element-caption\">Reactome human pathway over-representation for genes with joint baseline raw p&lt;0.001. The universe comprises all testable joint-model genes mapped to Reactome; pathways with 10\u2013500 background genes are tested. BH correction covers the entire eligible family, including zero overlaps. The 12 leading nominal pathways are shown with corrected q-values, overlap counts and fold enrichment. The gene set is unsigned and does not identify intelligence-increasing pathways.<\/figcaption><\/figure>\n\n<h2>Putting variants back on the protein<\/h2>\n<p>The interactive viewers are self-contained HTML downloads. Download a viewer from Dropbox and open it in a browser to rotate the reference structure and inspect the mapped variants.<\/p>\n<p>The 3D views connect association statistics with physical positions on the protein. Missense variants are placed at their mapped amino-acid residues in authentic AlphaFold reference models, with sequence and isoform identity checked before mapping. Color shows the sign of the variant\u2019s fluid-intelligence association; separate symbols and controls distinguish consequence class, nominal variant evidence, annotation values, and missing annotations.<\/p>\n<p>We include the primary set, the secondary FDR set, and the requested <span class=\"gene\">MYEF2<\/span> example, with the evidence category stated for each view. Low local confidence or uncertain positioning between domains limits structural interpretation; these displays do not establish a confidently folded global structure or a functional pocket.<\/p>\n<p>For <span class=\"gene\">KDM5B<\/span>, the main view uses the 1,580-residue UniProt isoform <strong>Q9UGL1-2<\/strong>, matching the source protein <code>ENSP00000356233.2<\/code> and transcript <code>ENST00000367264.7<\/code>. 613 of 630 baseline variants map after exact reference-amino-acid checks. A validated sequence crosswalk supports the separate 1,544-residue canonical view.<\/p><p>For <span class=\"gene\">MYEF2<\/span>, 216 of 220 variants map to the Q9P2K5 reference. Three splice variants and R426Q remain unmapped: the current model has Q426 where the source substitution expects R. These variants retain their association statistics and annotations in the tables.<\/p><p><span class=\"gene\">FBRSL1<\/span> uses the source-matched 1,045-residue Q9HCM7 model; 913\/922 baseline variants map. Its actual model confidence is low overall: mean C-alpha pLDDT 45.6, with 823\/1045 residues (78.8%) below 50. We therefore use it as a sequence-anchored location display and do not infer a confident global fold or functional pocket.<\/p><p><span class=\"gene\">ANKRD12<\/span> uses the source-matched Q6UB98 reference with 2,062 residues; 885\/890 baseline variants map. Mean C-alpha pLDDT is 41.15, with 77.1% of residues below 50. Five splice-only variants remain unmapped. This low-confidence model supports a reference-location display, not a confident global fold or functional pocket.<\/p>\n<p>In these views, a positive effect means <strong>associated with a higher score in this dataset<\/strong>. It does not establish a causal gain of function or an intelligence-enhancing mutation. AlphaMissense pathogenicity predicts the consequences of missense variation; it does not predict the direction of a fluid-intelligence association.<\/p>\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/xfugueprtnzbhamnrqgvb\/ANKRD12_3D_Atlas.png?rlkey=ihh6nfc53l2c1qa12f72fuvqb&amp;raw=1\" target=\"_blank\" rel=\"noopener\"><img src=\"https:\/\/www.dropbox.com\/scl\/fi\/xfugueprtnzbhamnrqgvb\/ANKRD12_3D_Atlas.png?rlkey=ihh6nfc53l2c1qa12f72fuvqb&amp;raw=1\" alt=\"ANKRD12: authentic reference coordinates and observed variant directions\" width=\"3520\" height=\"2200\" decoding=\"async\" loading=\"lazy\"><\/a><figcaption class=\"wp-element-caption\">ANKRD12 (ENSG00000101745); Primary planned-familywise association. Available-family BH values are conditional on the available source coverage. Reference positions are descriptive, with local confidence and mapping limits; positive beta is higher-score association, not causal enhancement, a mutant fold, or a spatial association test. Source-matched Q6UB98 canonical isoform, AlphaFold v6, 2,062 residues; protein ENSP00000262126.3, transcript ENST00000262126.9. 885\/890 baseline variants map and five splice-only variants remain unassigned. Mean C-alpha pLDDT 41.15; 77.1% of residues below 50. Global-fold and pocket inference are unsupported. Evidence label: Primary planned-familywise association; joint baseline p=5.143e-11, BH q=9.882e-07. BH q and evidence category are conditional on the available source coverage.<\/figcaption><\/figure>\n<p><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/5l3cwrei607genmcdwk9w\/3d_ENSG00000101745.html?rlkey=awmfl5j1fcuo6hxyhb583usx0&amp;dl=1\">Download ANKRD12 interactive 3D viewer (HTML)<\/a><\/p>\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/w31vp4ml5ibez4zhrp8wn\/KDM5B_3D_Atlas.png?rlkey=7rqnxf6rca436frxzq481ours&amp;raw=1\" target=\"_blank\" rel=\"noopener\"><img src=\"https:\/\/www.dropbox.com\/scl\/fi\/w31vp4ml5ibez4zhrp8wn\/KDM5B_3D_Atlas.png?rlkey=7rqnxf6rca436frxzq481ours&amp;raw=1\" alt=\"KDM5B: authentic reference coordinates and observed variant directions\" width=\"3520\" height=\"2200\" decoding=\"async\" loading=\"lazy\"><\/a><figcaption class=\"wp-element-caption\">KDM5B (ENSG00000117139); Primary planned-familywise association. Available-family BH values are conditional on the available source coverage. Reference positions are descriptive, with local confidence and mapping limits; positive beta is higher-score association, not causal enhancement, a mutant fold, or a spatial association test. Main view: source-matched Q9UGL1-2, AlphaFold v6, 1,580 residues; 613\/630 baseline variants map. Source protein ENSP00000356233.2, transcript ENST00000367264.7. The canonical 1,544-residue view uses an explicit validated sequence crosswalk. Evidence label: Primary planned-familywise association; joint baseline p=6.235e-07, BH q=0.00599. BH q and evidence category are conditional on the available source coverage.<\/figcaption><\/figure>\n<p><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/3e9sh10jhunuva57h806l\/3d_ENSG00000117139.html?rlkey=ydz1f5qyj8w0e2w306hrtf6it&amp;dl=1\">Download KDM5B interactive 3D viewer (HTML)<\/a><\/p>\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/6jo34osklvmntenexr52g\/FBRSL1_3D_Atlas.png?rlkey=1cyo0vv48h022h7srie55urc0&amp;raw=1\" target=\"_blank\" rel=\"noopener\"><img src=\"https:\/\/www.dropbox.com\/scl\/fi\/6jo34osklvmntenexr52g\/FBRSL1_3D_Atlas.png?rlkey=1cyo0vv48h022h7srie55urc0&amp;raw=1\" alt=\"FBRSL1: authentic reference coordinates and observed variant directions\" width=\"3520\" height=\"2200\" decoding=\"async\" loading=\"lazy\"><\/a><figcaption class=\"wp-element-caption\">FBRSL1 (ENSG00000112787); Secondary BH-FDR candidate. Available-family BH values are conditional on the available source coverage. Reference positions are descriptive, with local confidence and mapping limits; positive beta is higher-score association, not causal enhancement, a mutant fold, or a spatial association test. Q9HCM7, AlphaFold v6, source ENST00000434748.2\/ENSP00000396160.2; 913\/922 baseline variants map. 7 HC splice variants and 2 alternative-protein-accession records remain unmapped. Actual PDB C-alpha mean pLDDT 45.6; 823\/1045 residues (78.8%) are below 50. Global-fold and pocket interpretation are unsupported by this low-confidence model. Evidence label: Secondary BH-FDR candidate; primary planned-familywise threshold not passed; joint baseline p=3.912e-06, BH q=0.02506. BH q and evidence category are conditional on the available source coverage.<\/figcaption><\/figure>\n<p><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/94biyg6nyx34vu0eknmlo\/3d_ENSG00000112787.html?rlkey=tmz79h3xtu33s9ly6opv0w8mf&amp;dl=1\">Download FBRSL1 interactive 3D viewer (HTML)<\/a><\/p>\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/4h60dsthvt9zuezimmqk5\/MYEF2_3D_Atlas.png?rlkey=ck8yx6n2wlb8fioh3l91hmz10&amp;raw=1\" target=\"_blank\" rel=\"noopener\"><img src=\"https:\/\/www.dropbox.com\/scl\/fi\/4h60dsthvt9zuezimmqk5\/MYEF2_3D_Atlas.png?rlkey=ck8yx6n2wlb8fioh3l91hmz10&amp;raw=1\" alt=\"MYEF2: authentic reference coordinates and observed variant directions\" width=\"3520\" height=\"2200\" decoding=\"async\" loading=\"lazy\"><\/a><figcaption class=\"wp-element-caption\">MYEF2 (ENSG00000104177); Nominal MYEF2 case study. Available-family BH values are conditional on the available source coverage. Reference positions are descriptive, with local confidence and mapping limits; positive beta is higher-score association, not causal enhancement, a mutant fold, or a spatial association test. Q9P2K5, AlphaFold v6; 216\/220 variants map. Three splice variants and R426Q remain unmapped. The retained QC-flagged splice donor 15-48152233-C-A supplies 98.37% of descriptive observed-HC inverse-variance information; this is not its contribution to the source burden test. Evidence label: Requested nominal case study; primary and secondary criteria not passed; joint baseline p=4.246e-05, BH q=0.1243. BH q and evidence category are conditional on the available source coverage.<\/figcaption><\/figure>\n<p><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/3t0hrvzma1mune0zzsgoh\/3d_ENSG00000104177.html?rlkey=y6b25pkya9imy93ifxym64y0z&amp;dl=1\">Download MYEF2 interactive 3D viewer (HTML)<\/a><\/p>\n<p>Loss-of-function variants require a different visual interpretation. A stop-gain or frameshift can be anchored at its affected residue when the transcript mapping is unambiguous. That anchor marks the beginning of the change in the reference protein; it does not depict the folded structure of a truncated protein. Splice variants without a defensible residue mapping remain in the variant table. Model confidence and mapping coverage accompany every structural view.<\/p>\n<p>These maps are a way to inspect candidate mechanisms and plan experiments. Displaying several variants near one another does not constitute a statistical test of 3D clustering. The current association model uses annotations, not spatial proximity.<\/p>\n\n<h2>Keeping missing annotations visible<\/h2>\n<p>Missing annotations were imputed rather than used to discard a variant. Within each gene\u2019s AF \u2264 0.05 baseline set, a missing transformed HMM or AlphaMissense value receives the mean of the observed transformed values. Missing loss-of-function status receives the observed high-confidence-LoF proportion. If a gene has no observed values for an annotation, we use the phenotype-wide observed mean. Known labels remain unchanged.<\/p>\n<p>Imputation preserves association records but cannot replace missing annotation information. If a gene has no observed LoF labels, its imputed LoF column is constant, so a separate LoF coefficient cannot be identified. Recovering original source annotations can restore a varying moderator while leaving the variant effects and their uncertainties unchanged. Annotation coverage is therefore part of the evidence behind each gene\u2019s result.<\/p>\n<aside><strong>Annotation-coverage audit.<\/strong> In <span class=\"gene\">ANKRD12<\/span>, recovered source QC supplies 53 HC and 837 non-HC labels for the same 890 baseline variants. The LoF contrast becomes identifiable; before-and-after omnibus tests therefore have different tested alternatives. Source association estimates and HMM\/AlphaMissense values are preserved, and all 39 interval-QC flags remain included. The uniform recovery audit also verifies that all 14,860 previously cached gene-effect files are byte-identical before and after annotation recovery. <a href=\"https:\/\/www.dropbox.com\/scl\/fi\/myjk8xzynlc5vowvvh8l1\/ANKRD12_QC_Recovery.png?rlkey=p04e0wy86qddd1ig9ba3ac37j&amp;dl=1\">Download the annotation-recovery diagnostic<\/a>.<\/aside>\n<p>This procedure fills 601,693 HMM, 2,107,720 AlphaMissense, and 98,898 LoF entries in the baseline analysis. Imputed fractional LoF values are modeling values, not newly assigned biological LoF labels. Counts and direction summaries for observed LoF variants use known high-confidence labels only. The supplied HMM track covers autosomes; genes on X and Y remain included, with a constant imputed HMM term that cannot support a separate HMM coefficient.<\/p>\n<p>Interval and site QC flags do not exclude variants. Variant association <em>p<\/em>-values were not used to select the inputs, and no minimum allele-count filter was applied. The baseline contains 314,274 known interval-QC flags and 0 records with unknown interval-QC status. The downloadable tables and 3D tooltips retain these flags so that readers can inspect the evidence behind a result.<\/p>\n\n<h2>How much depends on allele frequency?<\/h2>\n<p>The primary analysis includes variants with AF \u2264 0.05. We also fit the same models at AF &lt; 0.001 and AF &lt; 0.0001, holding each gene\u2019s baseline imputation means fixed. These are sensitivity analyses; the baseline and two rarer subsets are not three independent replications.<\/p>\n<p>For <span class=\"gene\">ANKRD12<\/span>, the joint-test <em>p<\/em>-values are 5.14 \u00d7 10<sup>-11<\/sup> at AF \u2264 0.05, 2.24 \u00d7 10<sup>-11<\/sup> at AF &lt; 0.001, and 1.34 \u00d7 10<sup>-10<\/sup> at AF &lt; 0.0001. For <span class=\"gene\">KDM5B<\/span>, the joint-test <em>p<\/em>-values are 6.24 \u00d7 10<sup>-7<\/sup> at AF \u2264 0.05, 2.59 \u00d7 10<sup>-6<\/sup> at AF &lt; 0.001, and 6.34 \u00d7 10<sup>-6<\/sup> at AF &lt; 0.0001. For <span class=\"gene\">MYEF2<\/span>, the joint-test <em>p<\/em>-values are 4.25 \u00d7 10<sup>-5<\/sup> at AF \u2264 0.05, 0.0875 at AF &lt; 0.001, and 0.179 at AF &lt; 0.0001. A change across frequency thresholds can reflect a small number of influential variants, reduced information, or a different set of functional changes. It is useful evidence for follow-up, not a reason to combine the thresholds into one stronger claim.<\/p>\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/www.dropbox.com\/scl\/fi\/qaoit5d0r2fcn198a58zp\/Fluid_Intelligence_Blog_Sensitivity.png?rlkey=rh7i3g9xiihioa8c3fzv72m8x&amp;raw=1\" target=\"_blank\" rel=\"noopener\"><img src=\"https:\/\/www.dropbox.com\/scl\/fi\/qaoit5d0r2fcn198a58zp\/Fluid_Intelligence_Blog_Sensitivity.png?rlkey=rh7i3g9xiihioa8c3fzv72m8x&amp;raw=1\" alt=\"Sensitivity of gene associations to nested allele-frequency thresholds\" width=\"2586\" height=\"1929\" decoding=\"async\" loading=\"lazy\"><\/a><figcaption class=\"wp-element-caption\">Joint-model raw p-values and variant counts at three nested AF thresholds. Outlined cells pass the same planned 19,373-gene threshold at each AF cutoff. Row labels distinguish baseline FWER primary and 5% BH-FDR secondary genes; only the testable baseline joint-model family determines these labels. Rows include every baseline FWER\/BH-FDR gene, KDM5B, MYEF2, and up to eight strongest baseline tests. BH FDR uses only the baseline joint-model testable-gene family. Available-family BH classifications are conditional on the available source coverage. These overlapping subsets are sensitivities, not independent replications.<\/figcaption><\/figure>\n\n<h2>What would make an increasing-effect candidate convincing?<\/h2>\n<p>The search yields an uneven picture. Stronger evidence connects loss of function in ANKRD12 and KDM5B with lower scores. MYEF2 provides a positive nominal missense-class mean and a clear contrast with its LoF class, but its adjustment and synonymous comparisons limit an increasing-effect interpretation. FBRSL1 provides a positive LoF-direction candidate whose evidence depends strongly on the broader AF set. These distinctions are central to the search for genes and variants associated with higher fluid intelligence.<\/p>\n<p>Several limits remain. The public single-variant endpoints do not supply standard errors, so we reconstruct them as |\u03b2\u0302| \/ \u03a6<sup>\u22121<\/sup>(1 \u2212 <em>p<\/em>\/2), as requested. This is a normal-Wald reconstruction; it need not equal the original model-based uncertainty if the reported <em>p<\/em>-value came from a different calibration. The model treats variant errors as independent, because LD and carrier-overlap covariance are unavailable. It also treats imputed annotations as fixed and does not propagate their uncertainty. Retaining QC-flagged variants broadens inclusion and makes the recorded QC context particularly relevant when interpreting a candidate.<\/p>\n<div class=\"rivas-equation\"><p class=\"ql-center-displayed-equation\" style=\"line-height: 49px;\"><span class=\"ql-right-eqno\"> &nbsp; <\/span><span class=\"ql-left-eqno\"> &nbsp; <\/span><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/mrivas.su.domains\/gbe\/wp-content\/ql-cache\/quicklatex.com-7c9c3ab96151597b06849d47495e51b3_l3.png\" height=\"49\" width=\"168\" class=\"ql-img-displayed-equation quicklatex-auto-format\" alt=\"&#92;&#91;&#92;&#119;&#105;&#100;&#101;&#104;&#97;&#116;&#32;&#115;&#95;&#123;&#105;&#103;&#125;&#92;&#97;&#112;&#112;&#114;&#111;&#120;&#92;&#102;&#114;&#97;&#99;&#123;&#124;&#92;&#119;&#105;&#100;&#101;&#104;&#97;&#116;&#92;&#98;&#101;&#116;&#97;&#95;&#123;&#105;&#103;&#125;&#124;&#125;&#123;&#92;&#80;&#104;&#105;&#94;&#123;&#45;&#49;&#125;&#40;&#49;&#45;&#112;&#95;&#123;&#105;&#103;&#125;&#47;&#50;&#41;&#125;&#46;&#92;&#93;\" title=\"Rendered by QuickLaTeX.com\"\/><\/p><\/div>\n<p>These are exploratory analyses of an observational, population-based phenotype. A convincing increasing-effect candidate will require replication outside the discovery data, consistent directional evidence across appropriate sensitivity analyses, and functional experiments. Predicted pathogenicity alone cannot determine the direction of a fluid-intelligence effect. We make the statistics, annotation provenance and residue mappings available so those questions can be pursued directly.<\/p>\n<p>Explore the analysis: <a href=\"https:\/\/www.dropbox.com\/scl\/fi\/11b61dwq9eaqey33das3e\/Genebass_Fluid_Intelligence_Gene_Catalogue.tsv.gz?rlkey=rsadfmcqori37xiue045672pz&amp;dl=1\">Catalogue of all planned genes (TSV.gz)<\/a> \u00b7 <a href=\"https:\/\/www.dropbox.com\/scl\/fi\/sh6zmvu8ugpdvdzibf0jj\/gene_model_results.tsv.gz?rlkey=f3gl05xldmk5ia7ndw0aqpw2y&amp;dl=1\">Saved per-gene model tests (TSV.gz)<\/a> \u00b7 <a href=\"https:\/\/www.dropbox.com\/scl\/fi\/rqcroa66mrv1c6ylwmzxs\/gene_moderator_results.tsv.gz?rlkey=zexudf89tn8ottg30tbiwb6ez&amp;dl=1\">Saved moderator coefficients (TSV.gz)<\/a> \u00b7 <a href=\"https:\/\/www.dropbox.com\/scl\/fi\/tojg5muhcbrf9j8qpube1\/Unretrieved_Genes.tsv?rlkey=0r1ogc99fujzo87knoi66744s&amp;dl=1\">Unretrieved genes and coverage (TSV)<\/a> \u00b7 <a href=\"https:\/\/www.dropbox.com\/scl\/fi\/f33l5npy1opdj3gmahelf\/Interim_Code_Provenance.zip?rlkey=u0cgmq6d9cdpznyvpvuzd0flp&amp;dl=1\">Code and provenance (ZIP)<\/a> \u00b7 <a href=\"https:\/\/www.dropbox.com\/scl\/fi\/hb4np53t6kigdrgm8jsdk\/Reactome_pathway_enrichment.tsv?rlkey=4et8k208wq77hirv5qsdihmzu&amp;dl=1\">Reactome enrichment table<\/a> \u00b7 <a href=\"https:\/\/www.dropbox.com\/scl\/fi\/t6q0y1153qjyyz5xtufxi\/MYEF2_consequence_moderator_results.tsv?rlkey=upowtgq7rwr6npfxtuqfielrc&amp;dl=1\">MYEF2 consequence-class moderators<\/a>.<\/p>\n\n<h2>References<\/h2>\n<ol class=\"references\">\n<li id=\"ref-genebass\">Karczewski KJ, Solomonson M, Chao KR, et al. <em>Systematic single-variant and gene-based association testing of thousands of phenotypes in 394,841 UK Biobank exomes.<\/em> Cell Genomics. 2022;2:100168. <a href=\"https:\/\/doi.org\/10.1016\/j.xgen.2022.100168\">doi:10.1016\/j.xgen.2022.100168<\/a>. Genebass is the source resource; the fluid-intelligence analysis here uses the phenotype-specific sample of 128,302 participants.<\/li>\n<li id=\"ref-lauer\">Lauer L, Rivas MA. <em>Unified meta regression models for rare variant association studies.<\/em> bioRxiv. January 24, 2025. Preprint. <a href=\"https:\/\/doi.org\/10.1101\/2025.01.23.634522\">doi:10.1101\/2025.01.23.634522<\/a>.<\/li>\n<li id=\"ref-aguilar\">Aguilar O, Rivas M, Rivas MA. <em>A unified meta-regression model identifies genes associated with epilepsy.<\/em> medRxiv. Version 3, October 23, 2025. Preprint. <a href=\"https:\/\/www.medrxiv.org\/content\/10.1101\/2024.06.27.24309590v3\">doi:10.1101\/2024.06.27.24309590, v3<\/a>.<\/li>\n<li id=\"ref-palette\">Rivas MA, Beaudoin M, Gardet A, et al. <em>Deep resequencing of GWAS loci identifies independent rare variants associated with inflammatory bowel disease.<\/em> Nature Genetics. 2011;43:1066\u20131073. <a href=\"https:\/\/doi.org\/10.1038\/ng.952\">doi:10.1038\/ng.952<\/a>. Visual palette adapted from colors sampled from the original figures; the effect-direction and annotation color assignments in this article are new.<\/li>\n<li id=\"ref-alphamissense\">Cheng J, Novati G, Pan J, et al. <em>Accurate proteome-wide missense variant effect prediction with AlphaMissense.<\/em> Science. 2023;381:eadg7492. <a href=\"https:\/\/doi.org\/10.1126\/science.adg7492\">doi:10.1126\/science.adg7492<\/a>. AlphaMissense scores are from the public GRCh38 annotation release.<\/li>\n<li id=\"ref-alphafold\">Jumper J, Evans R, Pritzel A, et al. <em>Highly accurate protein structure prediction with AlphaFold.<\/em> Nature. 2021;596:583\u2013589. <a href=\"https:\/\/doi.org\/10.1038\/s41586-021-03819-2\">doi:10.1038\/s41586-021-03819-2<\/a>.<\/li>\n<li id=\"ref-alphafold-db\">Bertoni D, Tsenkov M, Magana P, et al. <em>AlphaFold Protein Structure Database 2025: a redesigned interface and updated structural coverage.<\/em> Nucleic Acids Research. 54:D358\u2013D362; published online November 22, 2025, January 2026 issue. <a href=\"https:\/\/doi.org\/10.1093\/nar\/gkaf1226\">doi:10.1093\/nar\/gkaf1226<\/a>. Structural coordinates are from the database\u2019s version-6 reference predictions; exact files and sequence checks are recorded with the viewers.<\/li>\n<li id=\"ref-chen\">Chen C-Y, Tian R, Ge T, et al. <em>The impact of rare protein coding genetic variation on adult cognitive function.<\/em> Nature Genetics. 2023;55:927\u2013938. <a href=\"https:\/\/doi.org\/10.1038\/s41588-023-01398-8\">doi:10.1038\/s41588-023-01398-8<\/a>.<\/li>\n<li id=\"ref-yamane\">Yamane K, Tateishi K, Klose RJ, et al. <em>PLU-1 is an H3K4 demethylase involved in transcriptional repression and breast cancer cell proliferation.<\/em> Molecular Cell. 2007;25:801\u2013812. <a href=\"https:\/\/doi.org\/10.1016\/j.molcel.2007.03.001\">doi:10.1016\/j.molcel.2007.03.001<\/a>.<\/li>\n<li id=\"ref-haas\">Haas S, Steplewski A, Siracusa LD, et al. <em>Identification of a sequence-specific single-stranded DNA binding protein that suppresses transcription of the mouse myelin basic protein gene.<\/em> Journal of Biological Chemistry. 1995;270:12503\u201312510. <a href=\"https:\/\/doi.org\/10.1074\/jbc.270.21.12503\">doi:10.1074\/jbc.270.21.12503<\/a>.<\/li>\n<li id=\"ref-wang\">Wang Y, et al. <em>PARP1-mediated PARylation activity is essential for oligodendroglial differentiation and CNS myelination.<\/em> Cell Reports. 2021;37:109695. <a href=\"https:\/\/doi.org\/10.1016\/j.celrep.2021.109695\">doi:10.1016\/j.celrep.2021.109695<\/a>.<\/li>\n<li id=\"ref-reactome\">Reactome Pathway Database. Release 98, human Ensembl gene-to-pathway mapping at all hierarchy levels; retrieved November 9, 2026. <a href=\"https:\/\/reactome.org\/download-data\">Source data<\/a>; <a href=\"https:\/\/reactome.org\/userguide\/analysis\">over-representation analysis documentation<\/a>. Exact source hashes, tested background and gene overlaps accompany this analysis.<\/li>\n<\/ol>\n<p class=\"small\">Analysis provenance: figures use the frozen analysis of 19,287 retrieved Genebass genes. The 86 source gaps remain documented, and available-family BH values are conditional on this coverage. Conceptual artwork is labeled separately from statistical and structure-based evidence.<\/p>\n\n<\/article>\n","protected":false},"excerpt":{"rendered":"<p>Positive genetic associations offer clues, but the route from a higher-score estimate to a convincing biological mechanism is difficult. MYEF2 shows both the promise and the uncertainty. Conceptual illustration of integrating constraint, missense pathogenicity and loss-of-function annotations. The ribbons are artistic motifs, not measured protein structures or statistical results. Can human genetic variation identify biological [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-703","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/mrivas.su.domains\/gbe\/wp-json\/wp\/v2\/posts\/703","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/mrivas.su.domains\/gbe\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mrivas.su.domains\/gbe\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mrivas.su.domains\/gbe\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/mrivas.su.domains\/gbe\/wp-json\/wp\/v2\/comments?post=703"}],"version-history":[{"count":5,"href":"https:\/\/mrivas.su.domains\/gbe\/wp-json\/wp\/v2\/posts\/703\/revisions"}],"predecessor-version":[{"id":709,"href":"https:\/\/mrivas.su.domains\/gbe\/wp-json\/wp\/v2\/posts\/703\/revisions\/709"}],"wp:attachment":[{"href":"https:\/\/mrivas.su.domains\/gbe\/wp-json\/wp\/v2\/media?parent=703"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mrivas.su.domains\/gbe\/wp-json\/wp\/v2\/categories?post=703"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mrivas.su.domains\/gbe\/wp-json\/wp\/v2\/tags?post=703"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}