Bash循环与条件判断：for/if/case全解

1
for bam in alignments/*.bam; do
2
    sample=$(basename "${bam}" .bam)
3

4
    # flagstat统计
5
    samtools flagstat "${bam}" > "stats/${sample}_flagstat.txt"
6

7
    # 深度统计
8
    samtools depth -a "${bam}" | \
9
        awk '{sum+=$3; count++} END {print "Mean depth:", sum/count}' \
10
        > "stats/${sample}_depth.txt"
11

12
    echo "${sample}: $(cat stats/${sample}_depth.txt)"
13
done

1
CHROMS=(chr1 chr2 chr3 chr4 chr5 chr6 chr7 chr8 chr9 chr10 \
2
        chr11 chr12 chr13 chr14 chr15 chr16 chr17 chr18 chr19 chr20 \
3
        chr21 chr22 chrX chrY chrM)
4

5
INPUT_VCF="merged_variants.vcf.gz"
6

7
for chrom in "${CHROMS[@]}"; do
8
    echo "Splitting ${chrom}..."
9

10
    bcftools view -r "${chrom}" "${INPUT_VCF}" \
11
        -o "by_chrom/${chrom}.vcf.gz" -O z
12

13
    bcftools index "by_chrom/${chrom}.vcf.gz"
14
done

1
files=(*.fastq.gz)
2
total=${#files[@]}
3
current=0
4

5
for f in "${files[@]}"; do
6
    ((current++))
7
    echo "[${current}/${total}] Processing ${f}..."
8

9
    fastp -i "${f}" -o "clean/${f}" -w 4
10
done
11

12
echo "All ${total} files processed!"

1
SAMPLES=(WT_1 WT_2 KO_1 KO_2)
2
TOOLS=(bwa bowtie2 hisat2)
3

4
for sample in "${SAMPLES[@]}"; do
5
    for tool in "${TOOLS[@]}"; do
6
        outdir="results/${tool}/${sample}"
7
        mkdir -p "${outdir}"
8

9
        echo "Aligning ${sample} with ${tool}..."
10
        # 不同工具用不同命令...
11
    done
12
done

1
# 读取样本列表
2
while IFS= read -r sample; do
3
    [[ -z "${sample}" || "${sample}" == \#* ]] && continue  # 跳过空行和注释
4
    echo "Processing: ${sample}"
5
    # ...处理逻辑
6
done < sample_list.txt

1
# ✗ 错误：管道中的while在子shell运行，变量不会传出来
2
total=0
3
cat counts.txt | while read -r n; do
4
    ((total += n))
5
done
6
echo "${total}"  # 输出：0 ——不是你要的！
7

8
# ✓ 正确：用重定向替代管道
9
total=0
10
while read -r n; do
11
    ((total += n))
12
done < counts.txt
13
echo "${total}"  # 输出：正确累加值

1
# 轮询等待文件生成
2
while true; do
3
    if [[ -f "pipeline_complete.flag" ]]; then
4
        echo "Pipeline finished!"
5
        break
6
    fi
7
    echo "Waiting..."
8
    sleep 60
9
done

1
# 等到磁盘空间够用
2
until [[ $(df /data --output=pcent | tail -1 | tr -d ' %') -lt 20 ]]; do
3
    echo "Disk usage still >20%, waiting..."
4
    sleep 300
5
done
6
echo "Disk OK, starting download"

1
MAPPING_RATE=$(samtools flagstat "${bam}" | \
2
    grep "mapped (" | grep -oP '\d+\.\d+(?=%)' | head -1)
3

4
if [[ $(echo "${MAPPING_RATE} < 70" | bc -l) -eq 1 ]]; then
5
    echo "WARNING: Low mapping rate (${MAPPING_RATE}%). Consider different aligner."
6
    # 或者发邮件通知
7
fi

1
READ_LENGTH=$(seqkit stats "${fastq}" | tail -1 | awk '{print $7}' | cut -d. -f1)
2

3
if [[ "${READ_LENGTH}" -lt 50 ]]; then
4
    ALIGNER="bowtie"        # 超短reads
5
elif [[ "${READ_LENGTH}" -lt 150 ]]; then
6
    ALIGNER="bwa"           # 短reads
7
elif [[ "${READ_LENGTH}" -lt 1000 ]]; then
8
    ALIGNER="minimap2 -x sr" # 中等长度
9
else
10
    ALIGNER="minimap2 -x map-ont"  # 长reads
11
fi
12

13
echo "Auto-selected aligner: ${ALIGNER}"

1
# 检查所有必需软件
2
check_tool() {
3
    command -v "$1" >/dev/null 2>&1 || {
4
        echo "ERROR: $1 not installed"
5
        exit 1
6
    }
7
}
8

9
for tool in bwa samtools bcftools fastp seqkit; do
10
    check_tool "${tool}"
11
done
12

13
echo "All tools available!"

1
SRA_ID="SRR12345678"
2

3
if [[ "${SRA_ID}" =~ ^(SRR|ERR|DRR)[0-9]{6,}$ ]]; then
4
    echo "Valid SRA ID: ${SRA_ID}"
5
else
6
    echo "ERROR: Invalid SRA ID format"
7
    exit 1
8
fi

1
# 运行比对
2
bwa mem -t 16 ref.fa reads.fq > aln.sam
3
ALN_EXIT=$?
4

5
if [[ ${ALN_EXIT} -eq 0 ]]; then
6
    echo "Alignment OK, sorting..."
7
    samtools sort -@ 8 aln.sam -o aln.bam
8
else
9
    echo "ERROR: Alignment failed with code ${ALN_EXIT}"
10
    exit ${ALN_EXIT}
11
fi

1
# ✓ 解决：先检查或用 shopt
2
shopt -s nullglob     # 没匹配时展开为空
3
for f in *.fastq.gz; do
4
    echo "Processing $f"
5
done
6
shopt -u nullglob     # 恢复默认
7

8
# 或者先判断
9
files=(*.fastq.gz)
10
if [[ ! -e "${files[0]}" ]]; then
11
    echo "No FASTQ files found"; exit 1
12
fi

1
# ✓ 用 || [[ -n "$line" ]] 兜底
2
while IFS= read -r line || [[ -n "$line" ]]; do
3
    echo "$line"
4
done < file.txt

1
# ✓ 用 [[ ]]（推荐）或加引号
2
[[ $a == "hello" ]]     # 内置，安全
3
[ "$a" == "hello" ]     # 传统，必须加引号

1
# ✓ 用 while read 或 find -print0 + while read -d ''
2
find . -name "*.bam" -print0 | while IFS= read -r -d '' f; do
3
    echo "Processing: ${f}"
4
done
5

6
# 或启用 globstar 替代 find
7
shopt -s globstar
8
for f in **/*.bam; do
9
    echo "Processing: ${f}"
10
done

1
# 如果必须用管道，用进程替换保持变量作用域
2
total=0
3
while read -r n; do
4
    ((total += n))
5
done < <(cat counts.txt)   # 进程替换，不是管道
6
echo "${total}"  # 正确

1
# 跳出多层用 break N
2
for i in {1..5}; do
3
    for j in {1..5}; do
4
        if [[ $i -eq 3 && $j -eq 3 ]]; then
5
            break 2  # 跳出两层
6
        fi
7
    done
8
done

1
set -u
2
for f in *.txt; do
3
    (( count++ ))        # 报错：count: unbound variable
4
done
5

6
# ✓ 先初始化
7
count=0
8
for f in *.txt; do
9
    (( count++ ))
10
done
11
echo "${count}"