> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.nvidia.com/clara/parabricks/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/clara/parabricks/_mcp/server. # 4.7.0-1 Release Notes **Highlights:** * New pangenome germline pipeline [pangenome\_germline](/tool-reference/tools/pangenome-germline) for GPU-accelerated pangenome alignment and variant calling. * Add support for NVIDIA GB10 (DGX Spark). * Added `--enable-small-model` support in [deepvariant](/tool-reference/tools/deepvariant) for shortread (WGS), PacBio, and ONT modes. * Updated version of [minimap2](/tool-reference/tools/minimap-2) to v2.30 along with some new presets, options and performance improvements. * Updated version of [giraffe](/tool-reference/tools/giraffe-vg-giraffe-gatk) to v1.70 along with performance improvements in single-ended mode. * Performance improvements for [rna\_fq2bam](/tool-reference/tools/rna-fq-2-bam). ## New Tools With Parabricks 4.7.0 we are releasing a GPU-accelerated [pangenome\_germline](/tool-reference/tools/pangenome-germline) pipeline. Pangenome\_germline is an end-to-end pipeline that integrates GPU-accelerated Giraffe with Pangenome-aware DeepVariant to streamline the workflow from FASTQ to VCF. ## Improvements All tools: * Added support for NVIDIA GB10 (DGX Spark). ### Tool Updates [fq2bam](/tool-reference/tools/fq-2-bam-bwa-mem-gatk) and [fq2bam\_meth](/tool-reference/tools/fq-2-bam-meth): * Improved performance on Hopper and Blackwell GPUs, with more modest gains on older architectures. * Allowed reads up to 512 bp to be processed on GPU, increasing the previous threshold from 500 bp. * When the `--max-read-length` filter is set above its default of 480 bp and up to 512 bp, all reads are processed on GPU. If the limit is set above 512 bp, reads longer than 512 bp are processed on CPU through the recovery mechanism. * Added support for manually setting the insert size distribution `-I` option from baseline BWA with `--bwa-options`. * For example: `--bwa-options="-I 250.54,20.25,500,20"`. The user interface is unchanged. * Added the `--cigar-on-gpu` option to offload CIGAR generation from the CPU (default) to the GPU; enabling this option may improve performance for CPU-bound runs. [fq2bam](/tool-reference/tools/fq-2-bam-bwa-mem-gatk), [fq2bam\_meth](/tool-reference/tools/fq-2-bam-meth), [giraffe](/tool-reference/tools/giraffe-vg-giraffe-gatk), [minimap2](/tool-reference/tools/minimap-2), [rna\_fq2bam](/tool-reference/tools/rna-fq-2-bam), [bamsort](/tool-reference/tools/bamsort), and associated pipelines: * Improved performance when using `--gpuwrite`. * Added a CPU recovery mode when using `--gpusort` to recover and continue processing even when the GPU may run out of memory. * GPU CRAM writer has reduced CPU memory requirements. Memory usage reductions can go up to 70% based on available host memory. [fq2bam](/tool-reference/tools/fq-2-bam-bwa-mem-gatk), [giraffe](/tool-reference/tools/giraffe-vg-giraffe-gatk), and [markdup](/tool-reference/tools/markdup): * Enabled read group id comparison in markdup single-ended mode. Can be disabled by passing `--ignore-rg-markdups-single-ended`. [deepvariant](/tool-reference/tools/deepvariant), [deepvariant\_germline](/tool-reference/tools/deepvariant-germline), [pacbio\_germline](/tool-reference/tools/pacbio-germline), and [ont\_germline](/tool-reference/tools/ont-germline): * Added support for the small model with `--enable-small-model`. This option is supported for shortread (WGS), PacBio, and ONT modes. * Improved host memory usage for long-read (PacBio and ONT) mode. [minimap2](/tool-reference/tools/minimap-2): * Updated version of minimap2 to v2.30. * Improved performance in PacBio mode and better GPU utilization. * Improved performance in splice presets with implementation of new alignment batching algorithm. * Added support for `--preset` options `lr:hq` and `splice:sr`. * Added argument `--copy-comment` for appending FASTQ comments to BAM output with auxiliary tag. * Added argument `--md` for outputting the MD tag. * Added argument `--both-strands` that forces minimap2 to consider both strands when finding canonical splicing sites GT-AG. [haplotypecaller](/tool-reference/tools/haplotypecaller): * Added `--exclude-intervals` (`-XL`) to specify genomic intervals to exclude from processing. [rna\_fq2bam](/tool-reference/tools/rna-fq-2-bam): * Auto mode: When `--num-streams-per-gpu` and/or `--num-threads` are omitted, their automatic configuration is triggered based on GPU and system memory. These settings can be overridden by providing explicit values. * Performance improvements from code optimizations. [giraffe](/tool-reference/tools/giraffe-vg-giraffe-gatk): * Updates to match the baseline VG Giraffe version v1.70. * Added GPU-accelerated Finalize Mappings (Dozeu) for single-end alignment. * Introduced `auto` mode for `--nstreams` that automatically configures the number of streams based on the GPU's device memory specifications. This provides sensible defaults for most systems but may still benefit from manual tuning for specific configurations. * Removed the `--minimizers-gpu-sort` option, as it is no longer needed. * Now used as the pangenome alignment stage in the new [pangenome\_germline](/tool-reference/tools/pangenome-germline) pipeline. ## Bug Fixes [mutectcaller](/tool-reference/tools/mutectcaller): * Added the missing read group (RG) header line in the normal sample BAM/CRAM output defined by `--mutect-bam-output`. [fq2bam](/tool-reference/tools/fq-2-bam-bwa-mem-gatk), [fq2bam\_meth](/tool-reference/tools/fq-2-bam-meth), [giraffe](/tool-reference/tools/giraffe-vg-giraffe-gatk), [minimap2](/tool-reference/tools/minimap-2), [markdup](/tool-reference/tools/markdup), [rna\_fq2bam](/tool-reference/tools/rna-fq-2-bam), and associated pipelines: * Addressed an issue preventing `--markdups-assume-sortorder-queryname` from taking effect when combined with `--gpusort`. [deepvariant](/tool-reference/tools/deepvariant), [deepsomatic](/tool-reference/tools/deepsomatic), and [pangenome\_aware\_deepvariant](/tool-reference/tools/pangenome-aware-deepvariant): * Fixed a potential GPU out-of-memory condition when `--num-streams-per-gpu` was set to auto (default). [rna\_fq2bam](/tool-reference/tools/rna-fq-2-bam): * Fixed missing alignments for reads with a high degree of multimapping. [fq2bam](/tool-reference/tools/fq-2-bam-bwa-mem-gatk) and [fq2bam\_meth](/tool-reference/tools/fq-2-bam-meth): * Fixed a memory leak that occurred during CPU recovery. * Fixed an issue where `--bwa-nstreams auto` (default value) and `--gpuwrite` could potentially run out of device memory with GPUs having more than 16 GB but less than 22 GB of memory. [fq2bam](/tool-reference/tools/fq-2-bam-bwa-mem-gatk), [fq2bam\_meth](/tool-reference/tools/fq-2-bam-meth), [giraffe](/tool-reference/tools/giraffe-vg-giraffe-gatk), [minimap2](/tool-reference/tools/minimap-2), [rna\_fq2bam](/tool-reference/tools/rna-fq-2-bam), [bamsort](/tool-reference/tools/bamsort), and associated pipelines: * Fixed a potential out-of-bounds device write when using `--gpuwrite` for accelerated BAM and CRAM writing. This risk was more likely to appear on ARM systems such as NVIDIA GB10. [fq2bam\_meth](/tool-reference/tools/fq-2-bam-meth): * Fixed `--preserve-file-symlinks` to work properly with methylation reference genome and indexes. [giraffe](/tool-reference/tools/giraffe-vg-giraffe-gatk): * Fixed a potential deadlock in paired-end mode that could occur when an input file had very few batches and all of them entered CPU recovery mode. [minimap2](/tool-reference/tools/minimap-2): * Fixed an issue where null padding could appear in output query names when using BAM input. For further information refer to the [Parabricks datasheet](https://resources.nvidia.com/en-us-hc-genomics/healthcare-genomics).