File Changes for angie

v394_preview2 to v394_base (2020-02-10 to 2020-02-17) v394

• src/hg/hgTracks/chainTrack.c
  • lines changed 2, context: html, text, full: html, text
    53cf1948cc8b9ea8a06f0285e2824e2a2b62353f Mon Feb 10 15:09:37 2020 -0800
    hgTracks: chain trackDb scoreFilter setting was ignored. refs #24695
• src/hg/makeDb/doc/hg19.txt
  • lines changed 39, context: html, text, full: html, text
    c11dda620b2fd111b46877fa8b05e7250a61b9f8 Tue Feb 11 12:50:33 2020 -0800
    Instructions for loading up NCBI's ReMap alignments as a chain track. refs #24449
• src/hg/makeDb/doc/hg38/hg38.txt
  • lines changed 208, context: html, text, full: html, text
    c1c71601b337d480f3e17ad3a1c72aac2b255352 Tue Feb 11 10:12:37 2020 -0800
    Rebuilt hg38 chainSelf: dropSelf with pslDropOverlap, fixed undetected errors that caused loss of chains, added -minScore=10000 filtering. refs #24695
  • lines changed 11, context: html, text, full: html, text
    8c68b963e315759727569797b7b32f2bc009a619 Tue Feb 11 11:25:10 2020 -0800
    oops, forgot download files for rebuilt hg38 Self Chain. refs #24695
  • lines changed 44, context: html, text, full: html, text
    c11dda620b2fd111b46877fa8b05e7250a61b9f8 Tue Feb 11 12:50:33 2020 -0800
    Instructions for loading up NCBI's ReMap alignments as a chain track. refs #24449
• src/hg/makeDb/trackDb/human/hg38/trackDb.ra
  • lines changed 1, context: html, text, full: html, text
    25d0ad24bdfdecb23813ac4ac751e8d7c8503ba7 Tue Feb 11 09:52:46 2020 -0800
    Add scoreFilter 20000 to hg38 chainSelf per Jim. refs #24695
• src/hg/mouseStuff/chainBridge/chainBridge.c
  • lines changed 140, context: html, text, full: html, text
    0e15377650ad1a2f32303ac3323b6a998b9ff937 Mon Feb 3 14:14:20 2020 -0800
    chainBridge: add -diffTolerance option; run bandExt in both directions in case best alignment is at end not start. refs #21650 When using chainBridge to extend through "gaps" in AGP correspondences, use large -diffTolerance, no need to require that t & q have similar sizes.
• src/hg/mouseStuff/chainBridge/tests/expected/someChange.bridged.chain
  • lines changed 4, context: html, text, full: html, text
    0e15377650ad1a2f32303ac3323b6a998b9ff937 Mon Feb 3 14:14:20 2020 -0800
    chainBridge: add -diffTolerance option; run bandExt in both directions in case best alignment is at end not start. refs #21650 When using chainBridge to extend through "gaps" in AGP correspondences, use large -diffTolerance, no need to require that t & q have similar sizes.
• src/hg/utils/automation/blastz-run-ucsc
  • lines changed 18, context: html, text, full: html, text
    a76cda6a69557a8c71016489147e4afd31266295 Mon Feb 3 15:19:13 2020 -0800
    blastz-run-ucsc: catch pipe errors; use pslDropOverlap for -dropSelf, for both lav and axt output. refs #24694, #24695 In MLQ #24587 a user reported some trivial self-alignments that should have been filtered - but when lastz produced axt output (for 2bit without suffix), we were not doing anything to filter those out. Now we use pslDropOverlap there. Worse, lavToAxt was failing when lastz output included double-sided gaps, and the failure was not propagated through the piped system() command, so undetected errors caused lots of alignments to be lost. So instead of using lavToAxt because of its -dropSelf option, use lavToPsl | pslDropOverlap.
• src/lib/gapCalc.c
  • lines changed 12, context: html, text, full: html, text
    5834a31cf2b9000f0e268647d47f03f6944f9e9d Fri Feb 7 13:44:20 2020 -0800
    Watch out for integer overflow for large gap penalties in gapCalcCost. refs #21650