File Changes for angie

v439_preview2 to v439_base (2022-10-24 to 2022-10-31) v439

• src/hg/hgPhyloPlace/hgPhyloPlace.c
  • lines changed 150, context: html, text, full: html, text
    0fffa3c31de4845a9bd3f06c0992f971e4d8a7a3 Fri Oct 28 15:08:06 2022 -0700
    Performance improvements for trees with millions of sequences: * Use @yceh's usher-sampled-server if configured; it preloads protobufs and can start placing sequences immediately using usher-sampled, a faster version of usher * Use usher-sampled instead of usher if server is not configured but usher-sampled is available * Load sample metadata file in a pthread while usher(-sampled(-server)) or matUtils is running * Skip checking for sample name clashes in uploaded fasta when using usher-sampled(-server)'s new --no-ignore-prefix option (but look for the prefix when parsing results) * Avoid parsing the protobuf and traversing the big tree unless absolutely necessary ** Subtrees from usher/matUtils have not included condensed nodes in a long time; remove lots of condensedNodes/summarization code from phyloPlace.c, runUsher.c, writeCustomTracks.c ** Use subtrees instead of big tree when possible (in findNearestNeighbor, treeToBaseAlleles, uploadedSamplesTree) ** Skip the informativeBases stuff that inhibits masking of sites from Problematic Sites set when the tree was built with an earlier version -- that pretty much never applies anymore now that only daily-updated trees are offered, not a range from old to new. ** Allow config.ra to specify a flat file of sample names (needed for searching user's uploaded names/IDs before calling matUtils) instead of getting names from the big tree
• src/hg/hgPhyloPlace/phyloPlace.c
  • lines changed 256, context: html, text, full: html, text
    0fffa3c31de4845a9bd3f06c0992f971e4d8a7a3 Fri Oct 28 15:08:06 2022 -0700
    Performance improvements for trees with millions of sequences: * Use @yceh's usher-sampled-server if configured; it preloads protobufs and can start placing sequences immediately using usher-sampled, a faster version of usher * Use usher-sampled instead of usher if server is not configured but usher-sampled is available * Load sample metadata file in a pthread while usher(-sampled(-server)) or matUtils is running * Skip checking for sample name clashes in uploaded fasta when using usher-sampled(-server)'s new --no-ignore-prefix option (but look for the prefix when parsing results) * Avoid parsing the protobuf and traversing the big tree unless absolutely necessary ** Subtrees from usher/matUtils have not included condensed nodes in a long time; remove lots of condensedNodes/summarization code from phyloPlace.c, runUsher.c, writeCustomTracks.c ** Use subtrees instead of big tree when possible (in findNearestNeighbor, treeToBaseAlleles, uploadedSamplesTree) ** Skip the informativeBases stuff that inhibits masking of sites from Problematic Sites set when the tree was built with an earlier version -- that pretty much never applies anymore now that only daily-updated trees are offered, not a range from old to new. ** Allow config.ra to specify a flat file of sample names (needed for searching user's uploaded names/IDs before calling matUtils) instead of getting names from the big tree
• src/hg/hgPhyloPlace/phyloPlace.h
  • lines changed 34, context: html, text, full: html, text
    0fffa3c31de4845a9bd3f06c0992f971e4d8a7a3 Fri Oct 28 15:08:06 2022 -0700
    Performance improvements for trees with millions of sequences: * Use @yceh's usher-sampled-server if configured; it preloads protobufs and can start placing sequences immediately using usher-sampled, a faster version of usher * Use usher-sampled instead of usher if server is not configured but usher-sampled is available * Load sample metadata file in a pthread while usher(-sampled(-server)) or matUtils is running * Skip checking for sample name clashes in uploaded fasta when using usher-sampled(-server)'s new --no-ignore-prefix option (but look for the prefix when parsing results) * Avoid parsing the protobuf and traversing the big tree unless absolutely necessary ** Subtrees from usher/matUtils have not included condensed nodes in a long time; remove lots of condensedNodes/summarization code from phyloPlace.c, runUsher.c, writeCustomTracks.c ** Use subtrees instead of big tree when possible (in findNearestNeighbor, treeToBaseAlleles, uploadedSamplesTree) ** Skip the informativeBases stuff that inhibits masking of sites from Problematic Sites set when the tree was built with an earlier version -- that pretty much never applies anymore now that only daily-updated trees are offered, not a range from old to new. ** Allow config.ra to specify a flat file of sample names (needed for searching user's uploaded names/IDs before calling matUtils) instead of getting names from the big tree
• src/hg/hgPhyloPlace/phyloPlaceMain.c
  • lines changed 9, context: html, text, full: html, text
    12ed3c2212ad565ac6ad2a713cf10ca455d5db27 Fri Oct 28 10:52:59 2022 -0700
    Add options -db and -protobuf for more flexible testing
• src/hg/hgPhyloPlace/runUsher.c
  • lines changed 842, context: html, text, full: html, text
    0fffa3c31de4845a9bd3f06c0992f971e4d8a7a3 Fri Oct 28 15:08:06 2022 -0700
    Performance improvements for trees with millions of sequences: * Use @yceh's usher-sampled-server if configured; it preloads protobufs and can start placing sequences immediately using usher-sampled, a faster version of usher * Use usher-sampled instead of usher if server is not configured but usher-sampled is available * Load sample metadata file in a pthread while usher(-sampled(-server)) or matUtils is running * Skip checking for sample name clashes in uploaded fasta when using usher-sampled(-server)'s new --no-ignore-prefix option (but look for the prefix when parsing results) * Avoid parsing the protobuf and traversing the big tree unless absolutely necessary ** Subtrees from usher/matUtils have not included condensed nodes in a long time; remove lots of condensedNodes/summarization code from phyloPlace.c, runUsher.c, writeCustomTracks.c ** Use subtrees instead of big tree when possible (in findNearestNeighbor, treeToBaseAlleles, uploadedSamplesTree) ** Skip the informativeBases stuff that inhibits masking of sites from Problematic Sites set when the tree was built with an earlier version -- that pretty much never applies anymore now that only daily-updated trees are offered, not a range from old to new. ** Allow config.ra to specify a flat file of sample names (needed for searching user's uploaded names/IDs before calling matUtils) instead of getting names from the big tree
• src/hg/hgPhyloPlace/vcfFromFasta.c
  • lines changed 14, context: html, text, full: html, text
    0fffa3c31de4845a9bd3f06c0992f971e4d8a7a3 Fri Oct 28 15:08:06 2022 -0700
    Performance improvements for trees with millions of sequences: * Use @yceh's usher-sampled-server if configured; it preloads protobufs and can start placing sequences immediately using usher-sampled, a faster version of usher * Use usher-sampled instead of usher if server is not configured but usher-sampled is available * Load sample metadata file in a pthread while usher(-sampled(-server)) or matUtils is running * Skip checking for sample name clashes in uploaded fasta when using usher-sampled(-server)'s new --no-ignore-prefix option (but look for the prefix when parsing results) * Avoid parsing the protobuf and traversing the big tree unless absolutely necessary ** Subtrees from usher/matUtils have not included condensed nodes in a long time; remove lots of condensedNodes/summarization code from phyloPlace.c, runUsher.c, writeCustomTracks.c ** Use subtrees instead of big tree when possible (in findNearestNeighbor, treeToBaseAlleles, uploadedSamplesTree) ** Skip the informativeBases stuff that inhibits masking of sites from Problematic Sites set when the tree was built with an earlier version -- that pretty much never applies anymore now that only daily-updated trees are offered, not a range from old to new. ** Allow config.ra to specify a flat file of sample names (needed for searching user's uploaded names/IDs before calling matUtils) instead of getting names from the big tree
• src/hg/hgPhyloPlace/writeCustomTracks.c
  • lines changed 35, context: html, text, full: html, text
    0fffa3c31de4845a9bd3f06c0992f971e4d8a7a3 Fri Oct 28 15:08:06 2022 -0700
    Performance improvements for trees with millions of sequences: * Use @yceh's usher-sampled-server if configured; it preloads protobufs and can start placing sequences immediately using usher-sampled, a faster version of usher * Use usher-sampled instead of usher if server is not configured but usher-sampled is available * Load sample metadata file in a pthread while usher(-sampled(-server)) or matUtils is running * Skip checking for sample name clashes in uploaded fasta when using usher-sampled(-server)'s new --no-ignore-prefix option (but look for the prefix when parsing results) * Avoid parsing the protobuf and traversing the big tree unless absolutely necessary ** Subtrees from usher/matUtils have not included condensed nodes in a long time; remove lots of condensedNodes/summarization code from phyloPlace.c, runUsher.c, writeCustomTracks.c ** Use subtrees instead of big tree when possible (in findNearestNeighbor, treeToBaseAlleles, uploadedSamplesTree) ** Skip the informativeBases stuff that inhibits masking of sites from Problematic Sites set when the tree was built with an earlier version -- that pretty much never applies anymore now that only daily-updated trees are offered, not a range from old to new. ** Allow config.ra to specify a flat file of sample names (needed for searching user's uploaded names/IDs before calling matUtils) instead of getting names from the big tree
• src/hg/utils/otto/sarscov2phylo/pango.clade-mutations.tsv
  • lines changed 126, context: html, text, full: html, text
    7e28b978133361b5bf86a3b0d13d41c83462e121 Mon Oct 24 09:34:04 2022 -0700
    Latest Cornelian deluge of lineages.
  • lines changed 6, context: html, text, full: html, text
    b7313108f102804030e287e76146971d9552c504 Fri Oct 28 15:18:29 2022 -0700
    A mere half-dozen new lineages.
• src/inc/pthreadWrap.h
  • lines changed 3, context: html, text, full: html, text
    b24c2a743b181481454816222301d0b832b5ccc9 Fri Oct 28 15:11:02 2022 -0700
    Add pthreadJoin for getting a thread worker function's return value.
• src/lib/pthreadWrap.c
  • lines changed 7, context: html, text, full: html, text
    b24c2a743b181481454816222301d0b832b5ccc9 Fri Oct 28 15:11:02 2022 -0700
    Add pthreadJoin for getting a thread worker function's return value.