PeakSnake/scripts/sample_from_bam.sh

#!/bin/bash

# 检查输入参数数量
if [ "$#" -ne 5 ]; then
    echo "Usage: $0 sample.bam chr1 chrM sample_name outputdir"
    exit 1
fi

# 读取输入参数
BAMFILE=$1
CHR1=$2
CHRM=$3
SAMPLENAME=$4
OUTPUTDIR=$5

# 创建输出目录
mkdir -p $OUTPUTDIR

# 生成临时文件的唯一标识符
TMPID=$(date +%s%N)  # 使用当前时间的纳秒作为唯一ID

# 提取 chr1 的 reads
samtools view -b $BAMFILE $CHR1 > $OUTPUTDIR/${SAMPLENAME}_${CHR1}_${TMPID}.bam

# 提取 chrM 的前 100 条 reads
samtools view -h $BAMFILE $CHRM | head -n 400 | samtools view -Sb - > $OUTPUTDIR/${SAMPLENAME}_${CHRM}_${TMPID}.bam

# 合并 chr1 和 chrM 的 reads
samtools merge -f $OUTPUTDIR/${SAMPLENAME}_merged_${TMPID}.bam $OUTPUTDIR/${SAMPLENAME}_${CHR1}_${TMPID}.bam $OUTPUTDIR/${SAMPLENAME}_${CHRM}_${TMPID}.bam

# 检查是单端还是双端测序
READFLAG=$(samtools view -f 1 -c $OUTPUTDIR/${SAMPLENAME}_merged_${TMPID}.bam)

# 根据测序类型生成对应的 FASTQ 文件
if [ $READFLAG -eq 0 ]; then
    # 单端
    samtools fastq $OUTPUTDIR/${SAMPLENAME}_merged_${TMPID}.bam > $OUTPUTDIR/${SAMPLENAME}_rep1_R1.fastq
    gzip $OUTPUTDIR/${SAMPLENAME}_rep1_R1.fastq
else
    # 双端
    samtools fastq -1 $OUTPUTDIR/${SAMPLENAME}_rep1_R1.fastq -2 $OUTPUTDIR/${SAMPLENAME}_rep1_R2.fastq $OUTPUTDIR/${SAMPLENAME}_merged_${TMPID}.bam
    gzip $OUTPUTDIR/${SAMPLENAME}_rep1_R1.fastq
    gzip $OUTPUTDIR/${SAMPLENAME}_rep1_R2.fastq
fi

# 清理临时文件
rm $OUTPUTDIR/${SAMPLENAME}_${CHR1}_${TMPID}.bam
rm $OUTPUTDIR/${SAMPLENAME}_${CHRM}_${TMPID}.bam
rm $OUTPUTDIR/${SAMPLENAME}_merged_${TMPID}.bam

echo "Completed."