File Changes for mspeir

v502_base to v503_preview (2026-08-10 to 2026-08-17) v503

• src/hg/makeDb/doc/hg38/singleCellSignalsPeaks.txt
  • lines changed 11, context: html, text, full: html, text
    fe02c55d26614f1d5059c884fb2594f9b4d5ad50 Wed Aug 12 08:02:19 2026 -0700
    singleCellSignalsPeaks: point at the hub build's new home, fix the file-copy script The hub build now lives in the cellBrowser repo under ucsc/allTracksHub, so the makeDocs link there instead of a personal work dir. HUB_BUILD in these scripts is the build's output dir, not its code; comments say so now, paths unchanged. copySingleCellSignalsPeaksFiles.py was silently copying nothing: it compared a whole line against "parent <composite>", but stanzas are now indented and read "parent <composite> off". Now dedents, matches the first token, and refuses to run on zero subtracks. Dry runs give 925 hg38 / 587 mm10, none missing. refs #37914
  • lines changed 1, context: html, text, full: html, text
    2162be65d294e930dfca338b0a3ef2a054449b36 Sat Aug 15 13:31:58 2026 -0700
    Make the singleCellSignalsPeaks metadata check mandatory, refs #38070 Missing facet metadata used to warn and write the .ra anyway, so a stanza file holding 22 of 925 subtracks produced a short .ra and exit 0. It is now an error, with --no-meta-check to skip it deliberately and --meta to point elsewhere. The metadata, palette and manifest are found relative to --stanzas instead of HUB_BUILD, so pointing --stanzas at another build no longer checks it against the default build's metadata. Moved the HUB_BUILD default off a personal hive directory to /hive/data/inside/cells/all-tracks-hub-build, beside the served hub. Both .ra files regenerate unchanged. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
• src/hg/makeDb/doc/mm10/singleCellSignalsPeaks.txt
  • lines changed 15, context: html, text, full: html, text
    fe02c55d26614f1d5059c884fb2594f9b4d5ad50 Wed Aug 12 08:02:19 2026 -0700
    singleCellSignalsPeaks: point at the hub build's new home, fix the file-copy script The hub build now lives in the cellBrowser repo under ucsc/allTracksHub, so the makeDocs link there instead of a personal work dir. HUB_BUILD in these scripts is the build's output dir, not its code; comments say so now, paths unchanged. copySingleCellSignalsPeaksFiles.py was silently copying nothing: it compared a whole line against "parent <composite>", but stanzas are now indented and read "parent <composite> off". Now dedents, matches the first token, and refuses to run on zero subtracks. Dry runs give 925 hg38 / 587 mm10, none missing. refs #37914
  • lines changed 1, context: html, text, full: html, text
    2162be65d294e930dfca338b0a3ef2a054449b36 Sat Aug 15 13:31:58 2026 -0700
    Make the singleCellSignalsPeaks metadata check mandatory, refs #38070 Missing facet metadata used to warn and write the .ra anyway, so a stanza file holding 22 of 925 subtracks produced a short .ra and exit 0. It is now an error, with --no-meta-check to skip it deliberately and --meta to point elsewhere. The metadata, palette and manifest are found relative to --stanzas instead of HUB_BUILD, so pointing --stanzas at another build no longer checks it against the default build's metadata. Moved the HUB_BUILD default off a personal hive directory to /hive/data/inside/cells/all-tracks-hub-build, beside the served hub. Both .ra files regenerate unchanged. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
• src/hg/makeDb/scripts/singleCellSignalsPeaks/copySingleCellSignalsPeaksFiles.py
  • lines changed 16, context: html, text, full: html, text
    fe02c55d26614f1d5059c884fb2594f9b4d5ad50 Wed Aug 12 08:02:19 2026 -0700
    singleCellSignalsPeaks: point at the hub build's new home, fix the file-copy script The hub build now lives in the cellBrowser repo under ucsc/allTracksHub, so the makeDocs link there instead of a personal work dir. HUB_BUILD in these scripts is the build's output dir, not its code; comments say so now, paths unchanged. copySingleCellSignalsPeaksFiles.py was silently copying nothing: it compared a whole line against "parent <composite>", but stanzas are now indented and read "parent <composite> off". Now dedents, matches the first token, and refuses to run on zero subtracks. Dry runs give 925 hg38 / 587 mm10, none missing. refs #37914
  • lines changed 1, context: html, text, full: html, text
    2162be65d294e930dfca338b0a3ef2a054449b36 Sat Aug 15 13:31:58 2026 -0700
    Make the singleCellSignalsPeaks metadata check mandatory, refs #38070 Missing facet metadata used to warn and write the .ra anyway, so a stanza file holding 22 of 925 subtracks produced a short .ra and exit 0. It is now an error, with --no-meta-check to skip it deliberately and --meta to point elsewhere. The metadata, palette and manifest are found relative to --stanzas instead of HUB_BUILD, so pointing --stanzas at another build no longer checks it against the default build's metadata. Moved the HUB_BUILD default off a personal hive directory to /hive/data/inside/cells/all-tracks-hub-build, beside the served hub. Both .ra files regenerate unchanged. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
• src/hg/makeDb/scripts/singleCellSignalsPeaks/makeSingleCellSignalsPeaksRa.py
  • lines changed 5, context: html, text, full: html, text
    fe02c55d26614f1d5059c884fb2594f9b4d5ad50 Wed Aug 12 08:02:19 2026 -0700
    singleCellSignalsPeaks: point at the hub build's new home, fix the file-copy script The hub build now lives in the cellBrowser repo under ucsc/allTracksHub, so the makeDocs link there instead of a personal work dir. HUB_BUILD in these scripts is the build's output dir, not its code; comments say so now, paths unchanged. copySingleCellSignalsPeaksFiles.py was silently copying nothing: it compared a whole line against "parent <composite>", but stanzas are now indented and read "parent <composite> off". Now dedents, matches the first token, and refuses to run on zero subtracks. Dry runs give 925 hg38 / 587 mm10, none missing. refs #37914
  • lines changed 0, context: html, text, full: html, text
    29afbb7595289b958dfb4fe966dc51c288d260cb Sat Aug 15 13:16:06 2026 -0700
    Count unrecognized G2P confidence values per record, not per output line, refs #38070 An HGNC ID can have several coordinate rows, so the old counter reported one record as many. Also fold case/whitespace so one new value is not logged as several, rename the remaining snake_case locals, and make makeSingleCellSignalsPeaksRa.py executable. Output is unchanged. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
  • lines changed 36, context: html, text, full: html, text
    2162be65d294e930dfca338b0a3ef2a054449b36 Sat Aug 15 13:31:58 2026 -0700
    Make the singleCellSignalsPeaks metadata check mandatory, refs #38070 Missing facet metadata used to warn and write the .ra anyway, so a stanza file holding 22 of 925 subtracks produced a short .ra and exit 0. It is now an error, with --no-meta-check to skip it deliberately and --meta to point elsewhere. The metadata, palette and manifest are found relative to --stanzas instead of HUB_BUILD, so pointing --stanzas at another build no longer checks it against the default build's metadata. Moved the HUB_BUILD default off a personal hive directory to /hive/data/inside/cells/all-tracks-hub-build, beside the served hub. Both .ra files regenerate unchanged. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
• src/hg/utils/otto/g2p/doG2p.py
  • lines changed 75, context: html, text, full: html, text
    29afbb7595289b958dfb4fe966dc51c288d260cb Sat Aug 15 13:16:06 2026 -0700
    Count unrecognized G2P confidence values per record, not per output line, refs #38070 An HGNC ID can have several coordinate rows, so the old counter reported one record as many. Also fold case/whitespace so one new value is not logged as several, rename the remaining snake_case locals, and make makeSingleCellSignalsPeaksRa.py executable. Output is unchanged. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
• src/hg/utils/otto/otto.crontab
  • lines changed 16, context: html, text, full: html, text
    4c06f1b66ceec8ba9e8f4618f632ccb271e2aa07 Wed Aug 12 15:15:20 2026 -0700
    Adding line for Cell Browser differential expression worker, and moving cbAnnotServer line up to be next to some other CB cron lines, refs #24912
  • lines changed 4, context: html, text, full: html, text
    87f71f1d645d6a2091dd4a21ea069c7b20922e0d Wed Aug 12 15:18:03 2026 -0700
    Fixing lines for Cell Browser differential expression worker, refs #24912