fe36c1b0cd47c32282b161d2f5abd89f19b168da gperez2 Wed Jul 29 15:27:29 2026 -0700 Made updates to the hg38 MPC track Description section, Display Conventions and Configuration section (added a legend table and other edits), and Methods section. Also updated gnomadMpc.ra to unhide transcript fields on the details page, reformat the mouseOver display, and clarify the Score field's scale, refs #37478 diff --git src/hg/makeDb/trackDb/human/hg38/gnomadMpc.html src/hg/makeDb/trackDb/human/hg38/gnomadMpc.html index 0d17cd3e560..69f6390e06d 100644 --- src/hg/makeDb/trackDb/human/hg38/gnomadMpc.html +++ src/hg/makeDb/trackDb/human/hg38/gnomadMpc.html @@ -1,141 +1,178 @@ <h2>Description</h2> <p> -Missense variants change a single amino acid in a protein and are a common -source of variants of uncertain significance (VUS): about 90% of missense -variants in ClinVar are VUS. The <b>MPC</b> score ("Missense deleteriousness -Prediction by Constraint") is a machine-learning score that flags missense -variants likely to be deleterious by combining three lines of evidence: -(i) regional missense constraint (how depleted the surrounding sub-genic -region is of rare missense variation in the general population), (ii) the -biochemical severity of the specific amino-acid substitution as captured by -PolyPhen-2, and (iii) cross-species conservation (phyloP). The model is -trained to separate pathogenic from benign missense variation under strong -heterozygous selection; higher scores indicate greater predicted -deleteriousness. The authors report that MPC ≥ 2.5 is strongly -enriched for de novo variants in individuals with severe developmental -disorders relative to their unaffected siblings, with MPC 2–2.5 showing -intermediate enrichment and MPC < 2 little enrichment. +This track shows MPC v4.1.1 ("Missense deleteriousness Prediction by +Constraint"), a machine-learning score that predicts which missense variants +are likely to be deleterious. It was computed by the Broad Institute gnomAD +team from the <b>gnomAD v4.1.1</b> release of 730,947 exomes aligned to +GRCh38/hg38. Scores are provided for every possible single-nucleotide +missense variant in 17,841 MANE Select or canonical protein-coding +transcripts that passed gnomAD QC, as well as for an additional 1,534 +transcripts that failed QC (the authors note that scores may be less +accurate in these 1,534 transcripts). </p> <p> -This track shows MPC v4.1.1, computed by the Broad Institute gnomAD team from -the <b>gnomAD v4.1.1</b> release of 730,947 exomes aligned to GRCh38. Scores -are provided for every possible single-nucleotide missense variant in 17,841 -MANE Select or canonical protein-coding transcripts that passed gnomAD QC, as -well as for an additional 1,534 transcripts that failed QC (the authors note -that scores may be less accurate in the latter). Our MPC track on hg19 -is an older release of the MPC score, calculated on gnomAD v2 (125,748 exomes) rather -than gnomAD v4 (730,947 exomes). +Missense variants change a single amino acid in a protein and are a common +source of variants of uncertain significance (VUS): about 90% of missense +variants in ClinVar are VUS. MPC predicts which missense variants are likely +deleterious by combining three lines of evidence: (i) regional missense constraint (how +depleted the surrounding sub-genic region is of rare missense variation in +the general population), (ii) the biochemical severity of the specific +amino-acid substitution and additional protein-structure and homology +information from PolyPhen-2, and (iii) cross-species conservation (phyloP). +Higher scores indicate greater predicted deleteriousness. </p> -<h2>Display Conventions and Configuration</h2> <p> -Two views of the same underlying data are available: +This data is available as two separate tracks: </p> <ul> -<li><b>gnomAD MPC</b> — four bigWig subtracks, one for each possible -alternate allele (A, C, G, T). At a given genomic position, the subtrack for -a given alternate allele shows the MPC score for the substitution from the -reference base to that alternate base. Positions where the alternate base -equals the reference base, or where no MPC score is available (for example, -outside coding regions), are drawn as 0. Scores range from 0 to 6, where 6 is -the ceiling assigned when a variant is more severe than every benign variant -in the MPC training set (the maximum real computed value is just over 5); -the track's default view limit is 0–3 because the vast majority of -variants score below 3. Mouseover shows the score. When a variant is -annotated against more than one transcript, the four bigWigs show the -<b>maximum</b> (most deleterious) MPC across transcripts.</li> -<li><b>gnomAD MPC overlaps</b> — a bigBed track restricted to the small +<li><b>gnomAD MPC</b> is a set of four bigWig subtracks, one for each +possible alternate allele (A, C, G, T), with scores ranging from 0 to 6. +At a given genomic position, the subtrack for a given alternate allele +shows the MPC score for the substitution from the reference base to that +allele. A score of 6 is the ceiling assigned when a variant is more severe +than every benign variant in the MPC training set (the maximum real +computed value is just over 5). When a variant is scored against more +than one transcript, the four bigWigs show the <b>maximum</b> (most +deleterious) MPC across transcripts.</li> +<li><b>gnomAD MPC overlaps</b> is a bigBed track covering the small subset of variants that are scored against more than one transcript (about -250,000, or 0.4% of the ~70 million scored variants). Each row carries the -full list of Ensembl transcripts scoring that variant together with their -individual MPC scores on the details page. Single-transcript variants are -not included in this track because the bigWig view already fully represents -them. A range filter on <tt>mpcMax</tt> is available from the track -configuration page. Items are colored by max MPC (gray < 1, -orange 1–3, red ≥ 3). Use this view to inspect transcripts whose -MPC scores disagree for the same underlying variant.</li> +250,000, or 0.4% of the ~70 million scored variants). Each item lists the +Ensembl transcripts and their individual MPC scores for that variant on +the details page, showing which transcripts' scores differ. Different +transcripts overlapping the same position don't always have an MPC score +for the same alternate alleles. Single-transcript variants are not +included in this track because the bigWig track already fully represents +them.</li> </ul> + <p> -Across the 250,000 multi-transcript variants, per-transcript MPC scores -typically agree within 0.5 units; only a few percent differ by more than -0.5. The bigBed view is the authoritative source for the full -transcript-level detail in those cases; the bigWig view collapses -transcripts by showing the maximum MPC at each position. +The <a href="../cgi-bin/hgTrackUi?db=hg19&g=gnomadMpc">MPC track on hg19</a> +is an older release of the MPC score, calculated on gnomAD v2 (125,748 +exomes) rather than gnomAD v4 (730,947 exomes). </p> +<h2>Display Conventions and Configuration</h2> +<ul> +<li><b>gnomAD MPC</b>: Each subtrack has a fixed color from the viridis +palette: A is purple, C is blue, G is teal, T is yellow. At a given +position, the subtrack matching the reference base shows 0, since that +isn't a real substitution. Positions outside the scored transcripts (for +example, outside coding regions) have no data. The track's default view +limit is 0–3 because the vast majority of variants score below 3. +Mouseover shows the score.</li> +<li><b>gnomAD MPC overlaps</b>: A range filter on <tt>mpcMax</tt> is +available from the track configuration page. Items are colored by max MPC: +<table> + <thead> + <tr> + <th style="border-bottom: 2px solid #6678B1;">Color</th> + <th style="border-bottom: 2px solid #6678B1;">Max MPC range</th> + </tr> + </thead> + <tr> + <th bgcolor="#C8C8C8"></th> + <th align="left">< 1</th> + </tr> + <tr> + <th bgcolor="#FFC864"></th> + <th align="left">1 – 2</th> + </tr> + <tr> + <th bgcolor="#FF9632"></th> + <th align="left">2 – 3</th> + </tr> + <tr> + <th bgcolor="#C80000"></th> + <th align="left">≥ 3</th> + </tr> +</table> +Mouseover shows the variant, its max MPC score, and the number of +transcripts that scored it.</li> +</ul> + <h2>Methods</h2> +<h3>Regional missense constraint (MCR)</h3> <p> -<b>Regional missense constraint (MCR).</b> For each of 17,841 QC-passing +For each of 17,841 QC-passing MANE Select or canonical coding transcripts, the authors tallied the observed rare missense variants (allele count > 0, allele frequency < 0.1%, %AN ≥ 20, QC PASS) in gnomAD v4.1.1 against the expected count under a position- and coverage-adjusted mutational model. A recursive likelihood-ratio test (Poisson model, p-value threshold 0.001, minimum 16 expected missense variants per sub-region) identifies change-points at which the transcript-wide observed/expected (OE) ratio deviates significantly; each resulting segment is a missense constraint region (MCR). 36% of transcripts (6,361/17,841) harbor two or more MCRs. MCR missense OE was calibrated against ClinVar P/LP vs. B/LB missense variants following ClinGen recommendations for the ACMG/AMP guidelines: OE ≤ 0.36 meets <i>moderate</i> evidence for pathogenicity, OE ≤ 0.59 meets <i>supporting</i> evidence for pathogenicity, OE > 0.97 and OE > 1.23 meet supporting and moderate evidence for benignity, respectively. </p> +<h3>MPC score</h3> <p> -<b>MPC score.</b> MPC is an XGBoost gradient-boosted-tree classifier that +MPC is an XGBoost gradient-boosted-tree classifier that takes as input (1) MCR missense OE, (2) gene-level constraint, (3) a per-substitution amino-acid severity feature, (4) the PolyPhen-2 -pathogenicity score, and (5) phyloP conservation. Training: 20,931 +pathogenicity score, and (5) phyloP conservation. The model is trained to +distinguish pathogenic missense variants under strong heterozygous +selection from benign missense variants under neutral to near-neutral +selection. Training: 20,931 "pathogenic" variants (high-quality ClinVar P/LP in 2,987 haploinsufficient genes with pHaplo ≥ 0.86 or in 359 non-LoF DD genes from Gene2Phenotype) vs. 93,638 "benign" variants (high-quality ClinVar B/LB or gnomAD variants with AF > 0.1% in the same gene set). The model is applied to all 70,313,598 possible exome-wide missense variants in the Ensembl VEP table. For a variant <i>i</i>, MPC is <i>d</i><sub><i>i</i></sub> = log<sub>10</sub>(<i>M</i> / <i>m</i><sub><i>i</i></sub>), where <i>M</i> is the number of benign training variants and <i>m</i><sub><i>i</i></sub> is the number of those with a fitted pathogenicity probability lower than variant <i>i</i>'s; when -<i>m</i><sub><i>i</i></sub> is 0 the score is capped at 6. Higher scores -indicate greater predicted deleteriousness. The authors caution that MPC is +<i>m</i><sub><i>i</i></sub> is 0 the score is capped at 6. Restricted to genes already +known to cause developmental disorders, the authors report that de novo +variants with MPC ≥ 2.5 are strongly enriched in affected individuals +relative to their unaffected siblings, with MPC 2–2.5 showing +intermediate enrichment and MPC < 2 little enrichment. The authors +caution that MPC is best suited to modelling strong fitness effects (as expected given its training set) and that naively taking the maximum of MPC and AlphaMissense <i>decreases</i> case/control discrimination for de novo variants relative to either score alone. Code for calculating the MPC scores and MCRs is available at the <a href="https://github.com/broadinstitute/regional_missense_constraint" target="_blank">broadinstitute/regional_missense_constraint</a> GitHub repository. </p> +<h3>UCSC Methods</h3> <p> -<b>At UCSC.</b> The precomputed MPC score table was downloaded from the +The precomputed MPC score table was downloaded from the gnomAD Broad public bucket at -<a href="https://storage.googleapis.com/gcp-public-data--gnomad/papers/2026-rmc/gnomad_v4.1.1_mpc.tsv.bgz">gs://gcp-public-data--gnomad/papers/2026-rmc/gnomad_v4.1.1_mpc.tsv.bgz</a>, +<a href="https://storage.googleapis.com/gcp-public-data--gnomad/papers/2026-rmc/gnomad_v4.1.1_mpc.tsv.bgz" target="_blank">gs://gcp-public-data--gnomad/papers/2026-rmc/gnomad_v4.1.1_mpc.tsv.bgz</a>, companion to the Hail-table release <tt>gnomad_v4.1.1_mpc.ht</tt> in the same directory. The input TSV contains one row per (locus, alleles, transcript) combination, for 70M rows. -Two Python scripts in <a target="_blank" -href="https://github.com/ucscGenomeBrowser/kent/tree/master/src/hg/makeDb/scripts/gnomadMpc">src/hg/makeDb/scripts/gnomadMpc</a> -convert to bigWig/bigBed formats. Build commands are documented -in the +href="https://github.com/ucscGenomeBrowser/kent/tree/master/src/hg/makeDb/scripts/gnomadMpc">Two Python scripts</a> +convert to bigWig/bigBed formats, available on GitHub. Build commands are +documented in the <a target="_blank" -href="https://github.com/ucscGenomeBrowser/kent/blob/master/src/hg/makeDb/doc/hg38/gnomadMpc.txt">hg38/gnomadMpc.txt</a> +href="https://github.com/ucscGenomeBrowser/kent/blob/master/src/hg/makeDb/doc/hg38/gnomadMpc.txt">gnomadMpc.txt</a> makeDoc file. </p> <h2>Data Access</h2> <p> The raw data can be explored interactively with the <a href="../cgi-bin/hgTables">Table Browser</a> or the <a href="../cgi-bin/hgIntegrator">Data Integrator</a>. For automated access, this track is available via our <a href="../goldenPath/help/api.html">API</a>. The underlying bigWig and bigBed files are at <a href="http://hgdownload.soe.ucsc.edu/gbdb/hg38/gnomAD/mpc/" target="_blank">our download server</a> as <tt>a.bw</tt>, <tt>c.bw</tt>, <tt>g.bw</tt>, <tt>t.bw</tt>, and <tt>mpcOverlaps.bb</tt>. Individual positions or whole chromosomes can be extracted with <tt>bigWigToBedGraph</tt> / <tt>bigWigToWig</tt> (for the bigWigs) or