my Salmon Snakemake file

snake.sh

#!/bin/bash
#
#SBATCH --job-name=snake
#SBATCH --time=240
#SBATCH --mem=1000

module load python
snakemake -j 4 --latency-wait 30 --cluster "sbatch -n 12 -N 1 --mem=10000 --time 60"

Snakefile

RUNS, = glob_wildcards("fastq/{run}_1.fastq.gz")

SALMON = "/proj/milovelab/bin/salmon-1.4.0_linux_x86_64/bin/salmon"

ANNO = "/proj/milovelab/anno"

rule all:
    input: expand("quants/{run}/quant.sf", run=RUNS)

rule salmon_index:
    input: "{ANNO}/gencode.vXYZ.transcripts.fa.gz"
    output: directory("{ANNO}/gencode.vXYZ-salmon_1.4.0")
    shell: "{SALMON} index --gencode -p 12 -t {input} -i {output}"

rule salmon_quant:
    input:
        r1 = "/pine/scr/m/i/milove/{sample}_1.fastq.gz",
        r2 = "/pine/scr/m/i/milove/{sample}_2.fastq.gz",
        index = "/proj/milovelab/anno/gencode.vXYZ-salmon_1.4.0"
    output:
        "quants/{sample}/quant.sf"
    params:
        dir = "quants/{sample}"
    shell:
        "{SALMON} quant -i {input.index} -l A -p 12 --gcBias "
        "--numGibbsSamples 20 --thinningFactor 100 "
        "-o {params.dir} -1 {input.r1} -2 {input.r2}"

Snakefile_with_QC

RUNS, = glob_wildcards("fastq/{run}_1.fastq.gz")

READS = ["1", "2"]

SALMON = "/proj/milovelab/bin/salmon-1.4.0_linux_x86_64/bin/salmon"

ANNO = "/proj/milovelab/anno"

rule all:
    input: "multiqc/multiqc_report.html"

rule salmon_index:
    input: "{ANNO}/gencode.v38.transcripts.fa.gz"
    output: directory("{ANNO}/gencode.v38-salmon_1.4.0")
    shell: "{SALMON} index --gencode -p 12 -t {input} -i {output}"

rule salmon_quant:
    input:
        r1 = "fastq/{sample}_1.fastq.gz",
        r2 = "fastq/{sample}_2.fastq.gz",
        index = "/proj/milovelab/anno/gencode.v38-salmon_1.4.0"
    output:
        "quants/{sample}/quant.sf"
    params:
        dir = "quants/{sample}"
    shell:
        "{SALMON} quant -i {input.index} -l A -p 12 --gcBias "
        "--numGibbsSamples 20 --thinningFactor 100 "
        "-o {params.dir} -1 {input.r1} -2 {input.r2}"

rule fastqc:
    input:
        "fastq/{sample}.fastq.gz"
    output:
        "qc/{sample}/{sample}_fastqc.html"
    params:
        dir = "qc/{sample}"
    shell:
        "fastqc --quiet -t 12 --outdir {params.dir} {input}"

rule multiqc:
    input:
        expand(["quants/{run}/quant.sf",
                "qc/{run}_{read}/{run}_{read}_fastqc.html"],
               run=RUNS, read=READS)
    output:
        "multiqc/multiqc_report.html"
    shell:
        "multiqc . -o multiqc"

zzz_download.R

scratch <- "/pine/scr/m/i/milove"

# see https://gist.github.com/mikelove/f539631f9e187a8931d34779436a1c01 for accession2url() definition

source("https://gist.githubusercontent.com/mikelove/f539631f9e187a8931d34779436a1c01/raw/6e6633aa5123358b70390ab738be1eef03a3df31/accession2url.R")

for (i in 1:nrow(x)) {
  print(paste("---",i,"---"))
  run <- x$Run[i]
  for (read in 1:2) {
    file <- paste0(run,"_",read,".fastq.gz")
    url <- file.path(accession2url(run), file)
    dest <- file.path(scratch, file)
    if (!file.exists(dest))
      download.file(url, dest)
  }
}