b40b82a00950316bb7c5eb1d92b2a6c68c503ab7 ann Thu Dec 8 14:02:21 2016 -0800 fixed typo diff --git src/hg/liftOver/README.txt src/hg/liftOver/README.txt index c09cb66..b280bcc 100644 --- src/hg/liftOver/README.txt +++ src/hg/liftOver/README.txt @@ -1,45 +1,45 @@ LICENSE The Genome Browser binaries and source code are freely available for academic, nonprofit, and personal use. A license is required for commercial download and installation of most Genome Browser binaries and source code, with the exception of specific items in some directories that are freely available for all uses. Please note the current -license infomation with the list of specific directory permissions: +license information with the list of specific directory permissions: http://genome.ucsc.edu/license/ ============================================================================= kent/src/hg/liftOver/ LiftOver has additional optional arguments for use with cross-species lifts: -multiple Allow multiple output regions -minSizeT, -minSizeQ Minimum chain size in target/query, when mapping to multiple output regions (default 0, 0) Depending on the size of the region being mapped, the species distance and the quality of the assembly, the user may need to adjust the minimum chain sizes. For ENCODE regions (.5 to 1.8 MB), I used -minSizeT=4000 and -minSizeQ=20000 for mammals except chimp, which, due to the fragmented assembly, I reduced minSizeQ to 10000 -minSizeQ=1000 for non-mammals (bird & fish) Also, for cross-species, they'll want to set the -minMatch very low (e.g. .01), instead of the default .95, which is a reasonable default for within species. Hmmm, sounds like this should be a README in our liftOver downloads dirs. Finally, there's an additional tool, liftOverMerge, which is useful when lifting large regions. I used it for ENCODE, but don't think it likely that users will need it. Kate