1f5afdd131d8f41f9dead82d52cb5a723e443437 jnavarr5 Fri Aug 28 15:25:59 2026 -0700 Announcing the EVE and popEVE tracks for hg38, refs #37791 #37763 diff --git src/hg/htdocs/goldenPath/newsarch.html src/hg/htdocs/goldenPath/newsarch.html index 604f23bb5cc..0752b27ad9d 100644 --- src/hg/htdocs/goldenPath/newsarch.html +++ src/hg/htdocs/goldenPath/newsarch.html @@ -52,30 +52,150 @@ <p>You can sign-up to get these announcements via our <a target=_blank href="https://groups.google.com/a/soe.ucsc.edu/g/genome-announce?hl=en">Genome-announce</a> email list. We send around one short announcement email every two weeks.</p> <p>Smaller software changes are not announced here. A summary of the three-weekly release changes can be found <a target=_blank href="https://genecats.gi.ucsc.edu/builds/versions.html">here</a>. For the full list of our daily code changes head to our <a href="https://github.com/ucscGenomeBrowser/kent/commits/master" target=_blank>GitHub page</a>. Lastly, see our <a href="credits.html" target="_blank"> credits page</a> for acknowledgments of the data we host.</p> <!-- ============= 2026 archived news ============= --> <a name="2026"></a> +<a name="082826"></a> +<h2>Aug. 28, 2026 Deleteriousness Predictions: popEVE and EVE for hg38</h2> +<p> +We are pleased to announce the release of two new proteome-wide missense variant effect +prediction tracks for hg38: +<a href="/cgi-bin/hgTrackUi?db=hg38&g=popEve&position=default" target="_blank">popEVE</a> and +<a href="/cgi-bin/hgTrackUi?db=hg38&g=eve&position=default" target="_blank">EVE</a>. +Both are deep generative models that predict variant pathogenicity by learning patterns of +natural sequence variation across species, without relying on clinical labels. +</p> + +<h3>EVE</h3> +<p> +<a href="https://evemodel.org" target="_blank">EVE</a> (Evolutionary model of Variant Effect) is a +deep generative model that predicts the pathogenicity of missense variants by learning patterns of +natural sequence variation across thousands of species, without relying on clinical labels. +Pre-computed scores are provided for all possible missense substitutions across 2,949 +disease-associated human proteins. +</p> + +<p> +Scores range from 0 to 1, with higher values indicating a greater predicted likelihood that a +variant is disease-relevant. As with any pathogenicity prediction score, EVE is intended as +supporting evidence rather than a stand-alone classifier. +</p> + +<p> +Each track entry spans one protein at its genomic locus. The heatmap columns correspond to +individual amino acid positions, placed at the codon's genomic coordinate, and the rows correspond +to the 20 standard amino acids, ordered by amino acid class. Each cell shows the EVE score for +substituting the wildtype amino acid at that position with the row amino acid. Empty cells +indicate the wildtype amino acid at a given position, or positions for which no score is +available. +</p> + +<div class="text-center" style="margin-top: 1.5em;"> +<a href="/cgi-bin/hgTracks?db=hg38&position=chr7%3A81752177-81752229&eve=pack" target="_blank"> +<img alt="Genome Browser screenshot of the EVE track at the HGF locus on hg38" +src="/images/newsArchImages/eve.png" width='55%'></a> +<p class="gbsCaption"><em>EVE missense variant effect scores for the HGF locus (chr7, +GRCh38/hg38). Each row is an amino acid substitution; each column is a protein position. Hovering +over a cell shows the substitution and EVE score.</em></p> +</div> + +<p> +Items in this track are colored according to score: +</p> +<ul> + <li><b>Red</b> – Likely pathogenic (score ≈ 1)</li> + <li><b>White</b> – Uncertain (score ≈ 0.5)</li> + <li><b>Blue</b> – Likely benign (score ≈ 0)</li> +</ul> + +<p> +<b>Note:</b> Zoom in to base level to see the full amino-acid heatmap with per-substitution scores +on mouseover. At wider zoom levels, the track switches to displaying a density graph. +</p> + +<h3>popEVE</h3> +<p> +<a href="https://pop.evemodel.org" target="_blank">popEVE</a> builds on EVE by combining its +evolutionary signal with predictions from the ESM-1v protein language model and human population +variation from the UK Biobank. A key difference is that EVE scores are calibrated separately +within each protein, whereas popEVE scores are calibrated across the entire proteome. That means a +popEVE score in one gene can be compared directly to a popEVE score in a completely different +gene. This is useful when prioritizing which of several candidate variants across different genes +is most likely to be damaging. The track provides popEVE scores for missense substitutions across +roughly 18,000 human proteins. +</p> + +<p> +popEVE scores are continuous, with lower, more negative values indicating greater predicted +deleteriousness. The authors define approximately −5.056 as a severe, high-confidence +deleterious threshold and −4.617 as a moderate threshold. Because the scores are calibrated +across the proteome, they can be compared directly between genes. +</p> + +<p> +Each track item represents a protein at its genomic locus and is displayed as a heatmap. Columns +correspond to amino acid positions, while rows represent possible amino acid substitutions. Each +cell shows the popEVE score for that substitution. +</p> + +<p> +Items are colored according to score: +</p> +<ul> + <li><b>Dark red</b> – Most deleterious (≤ −6.04)</li> + <li><b>Red-orange</b> – Severe (≈ −5.056)</li> + <li><b>Light orange</b> – Moderate (≈ −4.617)</li> + <li><b>White / very light</b> – Near the proteome-wide median (≈ −3.5)</li> + <li><b>Blue</b> – Most tolerated (≥ −2.41)</li> +</ul> + +<p> +Colors are interpolated between these values on a single proteome-wide gradient. As with other +computational variant-effect predictors, popEVE is intended to provide supporting evidence for +variant interpretation rather than serve as a stand-alone clinical classifier. +</p> + +<div class="text-center" style="margin-top: 1.5em;"> +<a href="/cgi-bin/hgTracks?db=hg38&position=chr3%3A38756746-38756842&popEve=pack" target="_blank"> +<img alt="Genome Browser screenshot of the popEVE track at the SCN10A locus on hg38" +src="/images/newsArchImages/popEve.png" width='55%'></a> +<p class="gbsCaption"><em>popEVE missense deleteriousness scores for all possible amino acid +substitutions in the SCN10A gene at chr3, GRCh38/hg38, where colors indicate predicted severity +from severe (red) to tolerated (blue).</em></p> +</div> + +<p> +We would like to thank Mafalda Dias, Jonathan Frazer, Debora S. Marks, Rose Orenbuch, and +colleagues at Harvard Medical School, the Centre for Genomic Regulation, and collaborating +institutions for developing popEVE and making the scores publicly available at +<a href="https://pop.evemodel.org" target="_blank">pop.evemodel.org</a>, and Jonathan Frazer, +Pascal Notin, Mafalda Dias, and Debora S. Marks at Harvard Medical School, along with Yarin Gal at +the University of Oxford, for making the EVE scores publicly available at +<a href="https://evemodel.org" target="_blank">evemodel.org</a>. Lou Nassar and Max Haeussler +developed these tracks, with QA by Jairo Navarro, Barali Kitiyakara, and Eliza Alde. +</p> + <a name="082626"></a> <h2>Aug. 26, 2026 gnomAD v4.1.1 and MPC tracks for hg38</h2> <p> We are excited to announce the updated <b>Genome Aggregation Database (gnomAD) v4.1.1 tracks</b> for human assembly hg38/GRCh38, and new <b>gnomAD Missense Deleteriousness Prediction by Constraint (MPC)</b> tracks, found in the <a target=_blank href="/cgi-bin/hgTrackUi?db=hg38&g=gnomadVariants">gnomAD superTrack</a>. The tracks are:</p> <ul> <li><b><a href="/cgi-bin/hgTrackUi?db=hg38&g=gnomadVariantsV4.1" target="_blank">gnomAD v4.1.1</a></b> (composite track): Shows exome and genome variants. This release revises the LOFTEE END_TRUNC GERP distance threshold from -58.0 to 0.0, reclassifying about 79,920 predicted loss-of-function variants from high-confidence to low-confidence.</li> <li><b><a href="/cgi-bin/hgTrackUi?db=hg38&g=gnomadPLI" target="_blank">gnomAD Constraint Metrics</a></b> (composite track): The v4.1.1 constraint metrics were recomputed using a Bayesian framework instead of the previous frequentist approach, expanded the coverage model