# Directory paths
DATA_DIR = "/home/debian/data1/breast/data/wes"
REFERENCE_DIR = "/home/debian/data2/genomes/human_GRCh38.p14"
REFERENCE_GENOME = os.path.join(REFERENCE_DIR, "GRCh38_latest_genomic.fna.gz")

samples = ['100B','101B']

rule bwa_index:
    input:
        REFERENCE_GENOME
    output:
        multiext(os.path.join(REFERENCE_DIR, "GRCh38_latest_genomic.fna.gz"), ".amb", ".ann", ".bwt", ".pac", ".sa")
    log:
        "logs/bwa_index.log"
    shell:
        """
        bwa index {input} &> {log}
        """

rule run_fastp:
    input:
        r1=os.path.join(DATA_DIR,{SAMPLE},"raw","{SAMPLE}-exome-tumor_{SEQ_ID}_L{LANE}_R1_001.fastq.gz",sample=SAMPLES),
        r2=os.path.join(DATA_DIR,{SAMPLE},"raw","{SAMPLE}-exome-tumor_{SEQ_ID}_L{LANE}_R2_001.fastq.gz",sample=SAMPLES)
    output:
        r1_trimmed="fastp/{SAMPLE}/{SAMPLE}-exome-tumor_{SEQ_ID}_L{LANE}_R1_001.fastq.gz",
        r2_trimmed="fastp/{SAMPLE}/{SAMPLE}-exome-tumor_{SEQ_ID}_L{LANE}_R1_001.fastq.gz"
    threads: 8
    shell:
        """
        fastp -i {input.r1} -I {input.r2} -o {output.r1_trimmed} -O {output.r2_trimmed} -w {threads}
        """

my files are like this:

250B-exome-tumor_S36_L001_R1_001.fastq.gz  250B-exome-tumor_S36_L002_R1_001.fastq.gz  250B-exome-tumor_S36_L003_R1_001.fastq.gz  250B-exome-tumor_S36_L004_R1_001.fastq.gz
250B-exome-tumor_S36_L001_R2_001.fastq.gz  250B-exome-tumor_S36_L002_R2_001.fastq.gz  250B-exome-tumor_S36_L003_R2_001.fastq.gz  250B-exome-tumor_S36_L004_R2_001.fastq.gz

And inside the wes directory I have a lot of directories, 100B/raw , 101B/raw , 102B/raw , and so on.

The problem is that the SEQ_ID is something unrelated to the sample number, so i have to find it first.
And that I have a variable number of LANES.

What’s the easiest way to deal with this inside snakemake?