Bash数组与字符串处理：参数展开、截取、替换

1
# 创建
2
samples=(WT_1 WT_2 KO_1 KO_2)
3
chroms=("chr1" "chr2" "chr3")
4
files=(*.bam)              # 通配符展开
5

6
# 访问
7
echo "${samples[0]}"       # 第一个元素：WT_1
8
echo "${samples[@]}"       # 所有元素（作为独立单词）
9
echo "${#samples[@]}"      # 数组长度：4
10
echo "${!samples[@]}"      # 所有索引：0 1 2 3
11

12
# 遍历
13
for s in "${samples[@]}"; do
14
    echo "Sample: ${s}"
15
done

1
# 追加
2
samples+=(WT_3 KO_3)       # 直接拼接
3
samples=("${samples[@]}" "extra")
4

5
# 按索引赋值
6
samples[0]="WT_CONTROL"
7

8
# 删除某个元素（实际是置空，不改变索引）
9
unset "samples[1]"
10

11
# 真正的删除+重新索引
12
samples=("${samples[@]}")  # 重建数组，跳过空位
13

14
# 切片
15
echo "${samples[@]:1:2}"   # 从索引1开始取2个
16

17
# 查找
18
for i in "${!samples[@]}"; do
19
    if [[ "${samples[$i]}" == "KO_2" ]]; then
20
        echo "Found KO_2 at index ${i}"
21
    fi
22
done

1
declare -A CONDITIONS
2
CONDITIONS=(
3
    [SRR100001]=treated
4
    [SRR100002]=treated
5
    [SRR100003]=control
6
    [SRR100004]=control
7
    [SRR100005]=treated
8
)
9

10
# 按条件分组
11
declare -A GROUPS
12
for sra in "${!CONDITIONS[@]}"; do
13
    cond="${CONDITIONS[$sra]}"
14
    GROUPS[$cond]="${GROUPS[$cond]:-} ${sra}"
15
done
16

17
# 输出分组
18
for cond in "${!GROUPS[@]}"; do
19
    echo "=== ${cond} ==="
20
    for sra in ${GROUPS[$cond]}; do
21
        echo "  ${sra}"
22
    done
23
done

1
declare -A GENE_COUNTS
2

3
# 从GTF中统计每个基因类型的出现次数
4
while IFS=$'\t' read -r chrom source feature start end score strand frame attrs; do
5
    if [[ "${feature}" == "gene" ]]; then
6
        # 从属性字段提取gene_type
7
        if [[ "${attrs}" =~ gene_type\ \"([^\"]+)\" ]]; then
8
            gene_type="${BASH_REMATCH[1]}"
9
            ((GENE_COUNTS[$gene_type]++))   # 关联数组做计数器
10
        fi
11
    fi
12
done < genes.gtf
13

14
# 输出统计
15
for gene_type in "${!GENE_COUNTS[@]}"; do
16
    echo "${gene_type}: ${GENE_COUNTS[$gene_type]}"
17
done | sort -t: -k2 -rn

1
path="/data/projects/RNASeq/results/sample1.bam"
2
filename="sample_WT_rep1_R1.fastq.gz"
3

4
# 删前缀 —— # 最短匹配，## 最长匹配
5
echo "${path#*/}"             # data/projects/RNASeq/results/sample1.bam
6
echo "${path##*/}"            # sample1.bam（basename的效果）
7

8
# 删后缀 —— % 最短匹配，%% 最长匹配
9
echo "${filename%.*}"         # sample_WT_rep1_R1.fastq
10
echo "${filename%%.*}"        # sample_WT_rep1_R1
11
echo "${filename%.fastq.gz}"  # sample_WT_rep1_R1
12

13
# 生信实战：提取样本ID
14
sample_id="${filename%%_R1*}"         # sample_WT_rep1
15
echo "Sample ID: ${sample_id}"

1
s="sample_WT_rep1_R1.fastq.gz"
2

3
# 首次替换
4
echo "${s/R1/R2}"                    # sample_WT_rep1_R2.fastq.gz
5

6
# 全部替换
7
echo "${s//r/R}"                     # sample_WT_Rep1_R1.fastq.gz
8

9
# 行首/行尾替换
10
echo "${s/#sample/SAMPLE}"           # SAMPLE_WT_rep1_R1.fastq.gz
11
echo "${s/%.gz/.bgz}"               # sample_WT_rep1_R1.fastq.bgz
12

13
# 生信实战：R1↔R2 配对
14
r1="sample_S1_L001_R1_001.fastq.gz"
15
r2="${r1/_R1_/_R2_}"                 # 最安全的替换方式
16
echo "${r2}"                         # sample_S1_L001_R2_001.fastq.gz