Major idea is that per-read haplotype likelihoods are now stored in a single unified object of class PerReadAlleleLikelihoodMap. Class implementation in theory hides internal storage details from outside work (still may need work cleaning up interface), and this object(or rather, a Map from Sample->perReadAlleleLikelihoodMap) is produced by UGCalcLikelihoods. The genotype calculation is also able to potentially use this info if needed. All InfoFieldAnnotations now get an extra argument with this map. Currently, this map is only produced for indels in UG, or for all variants within HaplotypeCaller. If this map is absent (SNPs in UG), the old Pileup interface is used, but it's avoided whenever possible. FORMAT annotations are not yet changed but will be focus of second step. Major benefit will be that annotations will be able to very easily discard non-informative reads for certain events. HaplotypeCaller also uses this new class, and no longer hard-codes the mapping of allele ->list(reads) but instead uses the same objects and interfaces as the rest of the modules. Code still needs further testing/cleaning/reviewing/debugging
-- Removed half-a*ssed attempt to automatically repair VCF files with bad headers, which allowed users to provide a replacement header overwriting the file's actually header on the fly. Not a good idea, really. Eric has promised to create a utility that walks through a VCF file and creates a meaningful header field based on the file's contents (if this ever becomes a priority)
-- Now possible to do -o /dev/stdout -bcf -l DEBUG > tmp.bcf and create a valid BCF2 file
-- Cleanup code to make sure extensions easier by moving to a setX model in VariantContextWriterStub
-- BCF2 is failing for some reason when merging tmp. files with parallel combine variants. ThreadLocalOutputTracker no longer sets deleteOnExit on the tmp file, as this prevents debugging. And it's unnecessary because each mergeInto was deleting files as appropriate
-- MergeInfo in VariantContextWriterStorage only deletes the intermediate output if an error occurs
-- All tests but one (using old bad VCF3 input) run unmodified with parallel code.
-- Disabled UNSAFE_VCF_PROCESSING for all but that test, which changes md5s because the output files have fixed headers
-- Minor optimizations to simpleMerge
-- BCF2 now determines whether it can safely write out raw genotype blocks, which is true in the case where the VCF header of the input is a complete, ordered subset of the output header. Added utilities to determine this and extensive unit tests (headerLinesAreOrderedConsistently)
-- Cleanup collapseStringList and exploreStringList for new unit tests of BCF2Utils. Fixed bug in edge case that never occurred in practice
-- VCFContigHeaderLine now provides its own key (VCFHeader.CONTIG_KEY) directly instead of requiring the user to provide it (and hoping its right)
-- More ways to access the data in VCFHeader
-- BCF2Writer uses a cache to avoid recomputing unnecessarily whether raw genotype blocks can be emitted directly into the output
-- Optimization of fullyDecodeAttributes -- attributes.size() is expensive and unnecessary. We just guess that on average we need ~10 elements for the attribute map
-- CombineVariants optimization -- filters are online HashSet but are sorted at the end by creating a TreeSet
-- makeCombinations is now makePermutations, and you can request to create the permutations with or without replacement
-- CombineVariants is now TreeReducible!
-- Integration tests running in parallel all pass except one (will fix) due to incorrect use of db=0 flag on input from old VCF format
-- Previous IO stub was hardcoded to write VCF. So when you ran -nt 2 -o my.bcf you actually created intermediate VCF files that were then encoded single threaded as BCF. Now we emit natively per thread BCF, and use the fast mergeInfo code to read BCF -> write BCF. Upcoming optimizations to avoid decoding genotype data unnecessarily will enable us to really quickly process BCF2 in parallel
-- VariantContextWriterStub forces BCF output for intermediate files
-- Nicer debug log message in BCF2Codec
-- Turn off debug logging of BCF2LazyGenotypesDecoder
-- BCF2FieldWriterManager now uses .debug not .info, so you won't see all of that field manager debugging info with BCF2 any longer
-- VariantContextWriterFactory.isBCFOutput now has version that accepts just a file path, not path + options
-- Expanded unit tests
-- Support for clean logging of results to logger
-- Refactored MyTime into AutoFormattingTime in Utils, out of TraversalEngine, for cleanliness and reuse
-- Added docs and contracts to StateMonitoringThreadFactory
-- GenomeLocParser cache was a major performance bottleneck in parallel GATK performance. With 10 thread > 50% of each thread's time was spent blocking on the MasterSequencingDictionary object. Made this a thread local variable.
-- Now we can run the GATK with 48 threads efficiently on GSA4!
-- Running -nt 1 => 75 minutes (didn't let is run all of the way through so likely would take longer)
-- Running -nt 24 => 3.81 minutes
-- The previously expanded ones are actually the missing values in the range. The previous ranges were correct. Removed the TODO to confirm them, as they are now officially confirmed
-- Includes header page
-- Table of arguments (Arguments)
-- Summary of counts (RecalData0)
-- Summary of counts by qual (RecalData1)
-- Fixed bug in output that resulted in covariates list always being null (updated md5s accordingly)
-- BQSR.R loads all relevant libaries now, include gplots, grid, and gsalib to run correctly
-- Added Write method to BCF2 types that directly converts int value to byte stream. Deleted writeRawBytes(int)
-- encodeTypeDescriptor semi-inlined into encodeType so that the tests for overflow are done in just one place
-- Faster implementation of determineIntegerType for int[] values
-- BCF2Type enum has an overloaded method to read the type as an int from an input stream. This gets rid of a case statement and replaces it with just minimum tiny methods that should be better optimized. As side effect of this optimization is an overall cleaner code organization
-- All low-level reads throw IOException instead of catching it directly. This allows us to not try/catch in readByte, improving performance by 5% or so
-- Optimize encodeTypeDescriptor with final variables. Avoid using Math.min instead do inline comparison
-- Inlined willOverflow directly in its single use
-- Old version converted doubles directly from strings. New version uses VariantContext getAttributeAsDouble() that looks at the values directly to determine how to convert from Object to Double (via Double.valueOf, (Double), or (Double)(Integer)).
-- getAttributeAsDouble() is now smart in converting integers to doubles as needed
-- Removed unnecessary logging info in BCF2Codec
-- Added integration tests to ensure that VQSR works end-to-end with BCF2 using sites version of the file khalid sent to me
-- Added vqsr.bcf_test.snps.unfiltered.bcf file for this integration test
-- Added bonferroni corrected p-value pruning, so you tell it how significant of a different you are willing to collapse in the tree, and it prunes the tree down to this maximum threshold
-- Penalty is now a phred-scaled p-value not the raw chi2 value
-- Split command line arguments in VisualizeContextTree into separate arguments for each type of pruning
-- Basically I was treating the context history in the wrong direction, effectively predicting the further bases in the context based on the closer one. Totally backward. Updated the code to build the tree in the right direction.
-- Added a few more useful outputs for analysis (minPenalty and maxPenalty)
-- Misc. cleanup of the code
-- Overall I'm not 100% certain this is even the right way to think about the problem. Clearly this is producing a reasonable output but the sum of chi2 values over the entire tree is just enormous. Perhaps a MCMC convergence / sampling criterion would be a better way to think about this problem?
-- Better output file name defaults
-- Fixed nasty bug where I included non-existant quals in the contexts to process because they showed up in the Cycle covariate
-- Data is processed in qual order now, so it's easier to see progress
-- Logger messages explaining where we are in the process
-- When in UPDATE mode we still write out the information for an equivalent prune by depth for post analysis
-- VisualizeContextTree now can write out an equivalent BQSR table determined after adaptive context merging of all RG x QUAL x CONTEXT trees
-- Docs, algorithm descriptions, etc so that it makes sense what's going on
-- VisualizeContextTree should really be simplified when into a single tool that just visualize the trees when / if we decide to make adaptive contexts standard part of BQSR
-- Misc. cleaning, organization of the code (recalibation tests were in private but corresponding actual files were public)
-- We are no likely to fail with an error when reading old BCF files, rather than just giving bad results
-- Added new class BCFVersion that consolidates all of the version management of BCF
-- Previous version would count all alt alleles as present in a sample, even if only 1 were present, because of the way VariantEval subsetted VCs
-- Updated code for subsetting VCs by sample to be clearer about how it handles rederiving alleles
-- Update a few pieces of code to get previous correct behavior
-- Updated a few MD5s as now ref calls at sites in dbSNP are counted as having a comp sites, and therefore show up in known sites when Novelty strat is on (which I think is correct)
-- Walkers that used old subsetting function with true are now using clearer version that does rederive alleles by default
Mark DePristo
Eric Banks
Guillermo del Angel
Khalid Shakir
Ami Levy Moonshine
Ryan Poplin
David Roazen
Joel Thibault