broadinstitute/picard

…rge of the gradle changes in htsjdk

Use htsjdk shadowJar instead of unmanaged libs

Adding a category and plot to WgsMetrics

…verage.

Adding a tool to compute wgs metrics for all sites as well as for only
non-zero coverage sites: CollectWgsMetricsWithNonZeroCoverage.  This
tool produces a plot that gets produced based on the fraction of the
genome covered at a given mean fold coverage.  This plots the normalized
(by the mean fold coverage) cumulative coverage for both all sites and
non-zero sites, with the latter useful for when we do not have
full-coverage over all sites.

Refactored CollectWgsMetrics to create a metrics collector.

…ar from picard.jar)

This reverts commit ac05c9c.

other metric collecting tools are called Collect*Metrics.

targeted metrics now filters out low quality bases and
enforces a minimum mapping quality, currently both set to the previous
defaults.

adding a median target coverage metric and the percentage of bases at 1x
or greater to the targeted metrics.

do not count overlapping bases in targeted metrics.  To do this, we need
to clone the SAMRecord, so adding an option to allow side effects when
computing the targeted metrics.  Also adding an option to *not* clip
overlapping reads when computing targeted metrics.

fixes a bug where we were not counting bases from supplemental records
in targeted metrics.  Supplemental records should have only been ignored
when counting metrics related to the number of records (not bases).

some metrics were computed a minimum coverage of two, which no longer
applies with high-quality capture technologies.  Updated to consider all
depths, including zero.

adding tests for CollectHsMetrics: overlapping reads, minimum base
quality, and minimum mapping quality.

adding fields to per target metrics: max_normalized_coverage,
min_coverage, max_coverage, and pct_target_bases_0x

- Fixed broken test due to change in order of iteration of HashSet/HashMap.

- Added Java 8 to the travis configuration but let the older JDKs there for now.

- Fixed failing test (SequencingArtifactsMetrics) that was failing due to change in HashMap order.

- Modified IlluminaBasecallsToSam to produce @RG lines with attributes in the same order as expected by the tests, and disabled one test that is fundamentally broken.

- Setting travis to JDK8 only.

- changed target in build.xml and removed some redundant <TYPES>

Added this CLP to the package-commands build target in build.xml