> 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.

# 4.5.1-1 Release Notes

**Highlights:**

* New features added in [rna\_fq2bam](/tool-reference/tools/rna-fq-2-bam): single-cell mode, quant mode and batch mode.
* `--mutect-f1r2-tar-gz` support for [mutectcaller](/tool-reference/tools/mutectcaller).
* Added support for haplotype sampling in [giraffe](/tool-reference/tools/giraffe-vg-giraffe-gatk), enabling personalized reference pangenome creation per sample.
* `vcf.gz` output support for all Parabricks variant callers.
* New options for RNA-seq data in [minimap2](/tool-reference/tools/minimap-2).
* Updated base container to CUDA 12.9.

## Improvements

### Tool Updates

[fq2bam](/tool-reference/tools/fq-2-bam-bwa-mem-gatk) and [fq2bam\_meth](/tool-reference/tools/fq-2-bam-meth):

* Optimized the performance of individual threads in the CPU thread pool during alignment.
* Additional optimizations for GPU performance.
* Improved error handling when using `--gpuwrite` option.

[rna\_fq2bam](/tool-reference/tools/rna-fq-2-bam):

* Added support for batch processing multiple input FASTQ files by repeatedly
  applying the `--in-fq` or `--in-se-fq` option.
* Added support for batch processing multiple input FASTQ files by passing in
  the `--in-fq-list` or `--in-se-fq-list` option, where the input
  is a text file containing a list of paired-end or single-end FASTQ files.
* Added support for single-cell sequencing mode, which can be enabled using the `--soloType Droplet` option along with additional necessary options including `--soloCBwhitelist`, `--soloFeatures`, `--soloStrand`, and `--soloBarcodeReadLength`.
* Added support for quantification mode using the `--quantMode` option.

[minimap2](/tool-reference/tools/minimap-2):

* Added support for RNA options `--minimizer-kmer-len` and
  `--forward-transcript-strand`, which are recommended for use in
  mapping long RNA-seq data in splice modes.

[mutectcaller](/tool-reference/tools/mutectcaller):

* Added argument `--mutect-f1r2-tar-gz` in mutectcaller to collect F1R2
  counts for FFPE samples.
* Added support for producing gzipped VCF files by specifying output file name with
  `.vcf.gz` extension.

[haplotypecaller](/tool-reference/tools/haplotypecaller):

* Added support for producing gzipped VCF files by specifying output file name with
  `.vcf.gz` extension.

[deepvariant](/tool-reference/tools/deepvariant):

* Added support for producing gzipped VCF files by specifying output file name with
  `.vcf.gz` extension.

[deepsomatic](/tool-reference/tools/deepsomatic):

* Added support for producing gzipped VCF files by specifying output file name with
  `.vcf.gz` extension.

[giraffe](/tool-reference/tools/giraffe-vg-giraffe-gatk):

* Added support for haplotype sampling, enabling personalized reference pangenome creation per sample.
* Added real-time CPU and host memory usage monitoring via `‑‑monitor‑usage` flag.

## Bug Fixes

[rna\_fq2bam](/tool-reference/tools/rna-fq-2-bam):

* The flag `--no-markdup` is correctly processed.

[fq2bam](/tool-reference/tools/fq-2-bam-bwa-mem-gatk), [fq2bam\_meth](/tool-reference/tools/fq-2-bam-meth), [rna\_fq2bam](/tool-reference/tools/rna-fq-2-bam), and [minimap2](/tool-reference/tools/minimap-2):

* Fixed a bug where the `--gpuwrite` option crashed with "invalid device ordinal" or "invalid configuration argument" errors in rare cases.

[collectmultiplemetrics](/tool-reference/tools/collectmultiplemetrics):

* Fixed a crash that occurred when the BAM file contained long reads.

* Fixed a crash that occurred when the BAM header was not compatible with the reference.

For further information refer to the
[Parabricks datasheet](https://resources.nvidia.com/en-us-hc-genomics/healthcare-genomics).