Skip to content

Latest commit

 

History

History
68 lines (46 loc) · 4.85 KB

analysisParameters.md

File metadata and controls

68 lines (46 loc) · 4.85 KB

Analysis parameters

All analysis parameters can be set in a runParams.yml file, which is then passed to nextflow with nextflow run -params-file runParams.yml. Alternatively each parameter in this file can also be set on the nextflow command-line directly, e.g. nextflow run --samples=samples.foo.csv

Note that nextflow options such as -resume or -params-file are given with a single -, while workflow parameters (e.g. --samples) are given with a double dash --. If both are used, command-line options overwrite values in the parameter file (e.g. runParams.yml).

Inputs

Sequencing reads

The alignment workflow can start from an Illumina sequencer run folder (bcl files), a directory with sequencing read files (fastq) or a directory with unaligned cram files processed on an Ultima sequencer. Specify one of these options:

  • runFolder : "path/to/runFolder"
  • fastqDir : "path/to/fastqs"
  • ultimaCramDir : "path/to/crams"

fastqDir is a directory containing all input fastq files for this analysis. See Fastq Generation for details on file names, etc.

runFolder is the top-level directory for a sequencer output (containing RunInfo.xml). The workflow uses Illumina bcl-convert for automatic fastq generation.

ultimaCramDir is a directory containing all input unaligned cram files pre-processed with Ultima trimmer.

Reporting only runs

If the sequencing data has been analyzed previously, these outputs can be re-used to generate updated reports. In this case previous workflow outputs are re-used with resultDir; see reportingNf.

Sample Information

  • samples : "samples.csv"

A file listing all samples in the analysis with their names, sample barcodes (RT) and optional sample settings

Reference Genome

  • genome : "/genomes/grch38/genome.json"

Path to a genome.json file that contains the location of all sequence and index files as well as other parameters for the reference genome to use.

Scale Bio RNA Kit Version

  • libStructure: "libQuantumV1.0.json"

This defines the version of the Scale Bio single-cell RNA kit used to generate the libraries. The default libQuantumV1.0.json matches version 1.0 of the Quantum ScaleRNA. For older Scale RNA (3 level) kits, set libStructure libV1.1.json (Scale RNA v1.1 kit and extended throughput kit).

Outputs

  • outDir: "ScaleRna.out" The name of the output directory for all analysis results. See outputs for a list of output files. Pre-existing files will be overwritten

BAM Output

By default bamOut and bcParserBamOut is set to false. bamOut being set to false suppresses alignment (.bam file) output from STAR. Gene expression results (.mtx), and all other workflow outputs, are still generated of course. bcParserBamOut controls publishing of per sample barcode demuxed unaligned bam files to the output directory. If alignments and sample demuxed files are not required for custom downstream analysis, disabling BAM output will save compute time and storage space.

Optional and Advanced Parameters

See nextflow.config for a list of all available parameters and their default values. The file also includes nextflow system options (compute resource requirements, etc.).

Multimapping Reads

The workflow uses STARsolo for alignments of reads to the genome and to transcripts. STARsolo handling of multimapping reads is controlled by these options

  • starMaxLoci: 6
    • Reads that map to more genomic locations that filtered out, regardless of transcriptome overlap.
  • starMulti: "PropUnique"
    • The algorithm used by STAR, to handle reads that match multiple genes, either because of genomic multimapping or because of overlapping gene annotations; See STARsolo documentation
  • roundCounts: "False"
    • If multi-gene reads are resolved, they lead to fractional unique transcript counts, e.g. "0.5" for a unique read that matches two genes equally well. If this setting is true, these counts are rounded to the closest full integer.

Cell Calling

See cellCalling.md for a description of the method used to call cell barcodes against background and related options.

Compute Resources

splitFastq, set to true by default, enables increased parallelization of the workflow by splitting the data based on input fastq files and RT barcodes. This can be set to false to reduce the number of compute jobs for a small analysis.

taskMaxMemory, taskMaxCpus and taskMaxTime control the maximum amount of compute resources a single task within the workflow is allowed to reserve. These can be lowered to match the available resources or limits of a users compute cluster.