1525618a814bcf6eb64c4147b99c520509a55d60
max
  Tue May 19 07:37:29 2026 -0700
Refresh mpraVarDB to 2026-05-19 snapshot; fix hg19 pos in name field.

Upstream landed the Mouri/Tewhey pvalue correction (5,092 rows with
pvalue > 1 -> 0) and replaced 47,156 placeholder fdr=1.0 values with
NaN. Rebuild from the new CSV; remove the temporary "pending upstream
fix" note in mpraVarDb.html.

Also closes the pre-existing "hg19 pos in non-rs name field" issue:
mpravardbToBed.py gained a post-liftOver step that rewrites the
chr:pos prefix inside non-rs names with the hg38 coordinates. 47,160
names were rewritten in this build (e.g. chr1:1403972:C>CG at hg38
chr1:1468591 now reads chr1:1468592:C>CG).

itemCount preserved at 239,028. refs #37359

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

diff --git src/hg/makeDb/doc/hg38/mpra.txt src/hg/makeDb/doc/hg38/mpra.txt
index abd93924541..1cc65f1fd1d 100644
--- src/hg/makeDb/doc/hg38/mpra.txt
+++ src/hg/makeDb/doc/hg38/mpra.txt
@@ -179,15 +179,43 @@
 #     hypercholesterolemia" (2,176 Kircher disease rows), "Alchol use
 #     disorder" -> "Alcohol use disorder" (88 Rao disease rows).
 #   - New fmt_mo() renders NaN floats as "NA" in the mouseOver helper
 #     fields rather than literal "nan"; 30,921 rows fixed.
 #   - Name + rsid handling tightened: a value is treated as an rsID
 #     only if it starts with "rs".  2,088 hg19-coord-style names like
 #     "1_1403972_CG" are now reformatted to "chr<X>:<hg38pos>:<ref>><alt>"
 #     and the rsid field is set to "" so the dbSNP linkout does not
 #     fire on a bogus value.
 #   - Removed the 250-char truncation that was cutting Griesemer
 #     descriptions mid-sentence; mpravardb.as switched the description
 #     and mpraStudy fields from "string" to "lstring" to allow full
 #     upstream text.
 #   - Pre-rebuild backup: mpravardb.bb.pre-2026-05-14-backup
 #   - itemCount preserved: 239,028.
+
+# =============================================================================
+# Snapshot refresh 2026-05-19 (Claude/max, RM #37359)
+# =============================================================================
+# Upstream MPRAVarDB published a refreshed CSV.  Schema unchanged; script
+# gained one post-liftOver step (fix_lifted_names) to rewrite the chr:pos
+# prefix inside non-rs name fields with the hg38 coordinates -- closes the
+# pre-existing "hg19 pos in name" issue from the prior makedoc entry.
+# 47,160 names were rewritten in this build.
+#
+# cd /hive/data/genomes/hg38/bed/mpra/mpravardb
+# wget 'https://mpravardb.rc.ufl.edu/session/4f77d030fa67160876b986a798875c6f/download/download_all?w=' -O mpravardb.csv
+# python3 ~/kent/src/hg/makeDb/scripts/mpravardb/mpravardbToBed.py
+#
+# Upstream changes vs. 2026-03-10 snapshot:
+#   - The 5,092 Mouri/Tewhey rows that previously stored a t-statistic in the
+#     pvalue field (pvalue > 1, max 8.96) are fixed: 0 rows with pvalue > 1
+#     remain.  Tao Wang's upstream correction has landed.  Accordingly the
+#     "Note (pending upstream fix)" paragraph was removed from mpraVarDb.html.
+#   - 47,156 rows that previously carried a placeholder fdr=1.0 now correctly
+#     report fdr=NaN (rendered "NA" via fmt_mo).  Net: rows with fdr=1.0
+#     dropped from 47,156 to 0; rows with fdr=NaN rose from 0 to 48,243.
+#   - Rows with pvalue=0 dropped from 7,398 to 7,151 (247 fewer literal-zero
+#     pvalues).
+# Pre-refresh backup: mpravardb.bb.pre-2026-05-19-backup
+# Input CSV rows: 242,818 (unchanged).  Lifted hg19->hg38: 209,899 of 210,013;
+# 114 unmapped.  Final itemCount: 239,028 (unchanged).  ~54,380 rows (~22%)
+# changed at least one column value vs. the prior bigBed.