#王振宇
#2021-5-27
#木聚糖缓解无纤维诱导紊乱

1.3.1 对所有.gz格式测序数据进行原文件夹解压

#! /bin/bash
#SBATCH --job-name=gzip
#SBATCH --partition=cast
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=8
#SBATCH --mem=16gb
#SBATCH --output=gzip_%j.log 
#SBATCH --qos normal

cleandata=~/WORKSPACE/rawdata/ff_xylan_pig/cleandata


#先从宏基因组样品命名信息中提取所有样品名
filename=$(cat ~/WORKSPACE/rawdata/ff_xylan_pig/cleandata/metadata.txt | tail -n+2 | awk '{print $1}')

cd ~/WORKSPACE/rawdata/ff_xylan_pig/cleandata

for i in $filename
        do
gzip -d ${cleandata}/${i}/*.gz 
done

#使用kneaddata对测序数据去宿主（猪）

#! /bin/bash
#SBATCH --job-name=kneaddata_filter
#SBATCH --partition=cast
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=72
#SBATCH --mem=144gb
#SBATCH --output=kneaddata_filter_%j.log 
#SBATCH --qos normal

module load jdk
module load bowtie2

filename=$(cat ~/WORKSPACE/rawdata/ff_xylan_pig/cleandata/metadata.txt | tail -n+2 | awk '{print $1}')

cleandata=~/WORKSPACE/rawdata/ff_xylan_pig/cleandata
pig_genome=~/WORKSPACE/database/pig_genome/bowtie_pig_reference_genome
ff_xylan_pig=~/WORKSPACE/result/ff_xylan_pig

    for i in $filename
        do
kneaddata --input ${cleandata}/${i}/${i}_1.clean.fq --input ${cleandata}/${i}/${i}_2.clean.fq \
          -db ${pig_genome}/pig_genome_11_1 \
          -p 72 --output ${ff_xylan_pig}/kneaddata_filter/ --bypass-trim --bypass-trf
      done

##metaphlan3物种注释
#! /bin/bash
#SBATCH --job-name=metaphlan3
#SBATCH --partition=cast
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=48
#SBATCH --mem=48gb
#SBATCH --output=metaphlan3_%j.log 
#SBATCH --qos normal

module load MetaPhlAn3
source $CONDA_SH
conda activate MetaPhlAn

metaphlan3=~/WORKSPACE/result/ff_xylan_pig/metaphlan3
filename=$(cat ~/WORKSPACE/rawdata/ff_xylan_pig/cleandata/metadata.txt | tail -n+2 | awk '{print $1}')
metaphlan3_database=~/WORKSPACE/database/metaphlan3_database
filterdata=~/WORKSPACE/rawdata/ff_xylan_pig/filterdata

cd ${metaphlan3}

  for i in $filename
  do
metaphlan ${rawdata}/${i}_1.clean_kneaddata_paired_1.fastq,${rawdata}/${i}_1.clean_kneaddata_paired_2.fastq --bowtie2out ${i}.bowtie2.bz2 --nproc 48 --input_type fastq -t rel_ab_w_read_stats -o ${metaphlan3}/absolutecount/${i}_ab_counts.txt --bowtie2db ${metaphlan3_database}
  done



#! /bin/bash
#SBATCH --job-name=megahit
#SBATCH --partition=cast
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=72
#SBATCH --mem=216gb
#SBATCH --output=megahit_%j.log 
#SBATCH --qos normal 

samplename=~/WORKSPACE/rawdata/ff_xylan_pig/filterdata
megahit=~/WORKSPACE/result/ff_xylan_pig/megahit

time megahit -t 72 \
             -1 `cat ${samplename}/samplename_XY_modify.txt | grep -E '4' | sed 's/^/WORKSPACE\/rawdata\/ff_xylan_pig\/filterdata\//;s/$/_1.clean_kneaddata_paired_1.fastq/' | tr '\n' ',' | sed 's/,$//'` \
             -2 `cat ${samplename}/samplename_XY_modify.txt | grep -E '4' | sed 's/^/WORKSPACE\/rawdata\/ff_xylan_pig\/filterdata\//;s/$/_1.clean_kneaddata_paired_2.fastq/' | tr '\n' ',' | sed 's/,$//'` \
             -o ${megahit} --min-contig-len 300 --min-count 2

time megahit -t 72 \
             -1 `cat ${samplename}/samplename_FF_modify.txt | grep -E '1' | sed 's/^/WORKSPACE\/rawdata\/ff_xylan_pig\/filterdata\//;s/$/_1.clean_kneaddata_paired_1.fastq/' | tr '\n' ',' | sed 's/,$//'` \
             -2 `cat ${samplename}/samplename_FF_modify.txt | grep -E '1' | sed 's/^/WORKSPACE\/rawdata\/ff_xylan_pig\/filterdata\//;s/$/_1.clean_kneaddata_paired_2.fastq/' | tr '\n' ',' | sed 's/,$//'` \
             -o ${megahit}/FF --min-contig-len 300 --min-count 2

##合并组装结果

cd ~/WORKSPACE/result/ff_xylan_pig/megahit

cat *.fa > merge_contig.fa

seqkit rename merge_contig.fa > unique_contig.fa

#1.5基因预测
#! /bin/bash
#SBATCH --job-name=prodigal
#SBATCH --partition=cast
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=72
#SBATCH --mem=72gb
#SBATCH --output=prodigal_%j.log 
#SBATCH --qos normal

megahit=~/WORKSPACE/result/ff_xylan_pig/megahit
prodigal=~/WORKSPACE/result/ff_xylan_pig/prodigal

time prodigal -i ${megahit}/merge_contig_unique.fa \
        -d ${prodigal}/gene.fa \
        -o ${prodigal}/gene.gff \
        -a ${prodigal}/protein.faa
        -p meta -f gff 

# 统计基因数量,6709811
    grep -c '>' ~/WORKSPACE/result/ff_xylan_pig/prodigal/gene.fa  

#完整基因，2886465
    grep -c 'partial=00' ~/WORKSPACE/result/ff_xylan_pig/prodigal/gene.fa  

#提取完整基因ID
    grep 'partial=00' ~/WORKSPACE/result/ff_xylan_pig/prodigal/gene.fa | cut -f1 -d ' '| sed 's/>//' > ~/WORKSPACE/result/ff_xylan_pig/prodigal/full_length.id

##根据完整基因id过滤基因集
    seqkit grep -f ~/WORKSPACE/result/ff_xylan_pig/prodigal/full_length.id ~/WORKSPACE/result/ff_xylan_pig/prodigal/gene.fa > ~/WORKSPACE/result/ff_xylan_pig/prodigal/full_length.fa

##完整基因集基本统计信息
    seqkit stat ~/WORKSPACE/result/ff_xylan_pig/prodigal/full_length.fa


transeq -sequence ./full_length.fa -outseq ./full_length_protein.fa -trim Y -clean

# 基因去冗余
# aS覆盖度，c相似度，G局部比对，g最优解，T多线程，M内存0不限制
# 2万基因2m，2千万需要2000h，可多线程加速
#! /bin/bash
#SBATCH --job-name=cd-hit-est
#SBATCH --partition=cast
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=72
#SBATCH --mem=72gb
#SBATCH --output=cd-hit-est_%j.log 
#SBATCH --qos normal

prodigal=~/WORKSPACE/result/ff_xylan_pig/prodigal
cd_hit=~/WORKSPACE/result/ff_xylan_pig/cd_hit

time cd-hit -i ${prodigal}/full_length_protein.fa \
        -o ${cd_hit}/cluster_protein_90.fa \
        -aS 0.9 -c 0.90 -G 0 -g 0 -T 36 -M 0

##去掉后缀_1
 sed -i 's/_1 / /' ~/WORKSPACE/result/ff_xylan_pig/cd_hit/cluster_protein_90.fa

##提取非冗余基因集（蛋白序列）id
grep "^>" ~/WORKSPACE/result/ff_xylan_pig/cd_hit/cluster_protein_90.fa | cut -d '#' -f1 | sed 's/>//' > id.txt

##根据提取的id过滤核酸序列
seqtk subseq ~/WORKSPACE/result/ff_xylan_pig/prodigal/full_length.fa id.txt > ~/WORKSPACE/result/ff_xylan_pig/cd_hit/cluster_nucleotide_90.fa

#1.7基因丰度计算

# 建索引, -t序列, -i 索引，10s
#! /bin/bash
#SBATCH --job-name=salmon
#SBATCH --partition=cast
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=36
#SBATCH --mem=72gb
#SBATCH --output=salmon_%j.log 
#SBATCH --qos normal

mkdir ~/WORKSPACE/result/ff_xylan_pig/salmon

cd_hit=~/WORKSPACE/result/ff_xylan_pig/cd_hit
salmon=~/WORKSPACE/result/ff_xylan_pig/salmon

time salmon index -t ${cd_hit}/cluster_nucleotide_90.fa -p 36 \
        -i ${salmon}/index

##基因定量
#! /bin/bash
#SBATCH --job-name=salmon
#SBATCH --partition=cast
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=36
#SBATCH --mem=72gb
#SBATCH --output=salmon_%j.log 
#SBATCH --qos normal


filename=$(cat ~/WORKSPACE/rawdata/ff_xylan_pig/cleandata/metadata.txt | tail -n+2 | awk '{print $1}')
index=~/WORKSPACE/result/ff_xylan_result/salmon/index
filterdata=~/WORKSPACE/rawdata/ff_xylan_pig/filterdata

for i in $filename
do
salmon quant \
        -i ${index} -l A -p 48 --meta \
        -1 ${filterdata}/${i}_1.clean_kneaddata_paired_1.fastq \
        -2 ${filterdata}/${i}_1.clean_kneaddata_paired_2.fastq \
        -o ~/WORKSPACE/result/ff_xylan_pig/salmon/${i}.quant
done

##基因定量结果合并
mkdir ~/WORKSPACE/result/ff_xylan_pig/salmon/result

##结果为相对丰度
salmon quantmerge  --quants ~/WORKSPACE/result/ff_xylan_pig/salmon/*.quant -o ~/WORKSPACE/result/ff_xylan_pig/salmon/result/gene.TPM

##结果为基因count数
salmon quantmerge  --quants ~/WORKSPACE/result/ff_xylan_pig/salmon/*.quant --column NumReads -o ~/WORKSPACE/result/ff_xylan_pig/salmon/result/gene.count



##dbCAN2碳水化合物功能注释
##构建索引
time diamond makedb \
      --in /home/wangjunjun/WORKSPACE/database/dbCAN_database/CAZyDB.07312020.fa \
      --db CAZyDB.07312020

##功能注释
#! /bin/bash
#SBATCH --job-name=dbCAN_diamond
#SBATCH --partition=cast
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=36
#SBATCH --mem=72gb
#SBATCH --output=dbCAN_diamond_%j.log 
#SBATCH --qos normal
dbCAN=~/WORKSPACE/database/dbCAN_database
cd_hit=~/WORKSPACE/result/ff_xylan_pig/cd_hit

mkdir ~/WORKSPACE/result/ff_xylan_pig/dbcan2

time diamond blastp --db ${dbCAN}/CAZyDB.07312018 --query ${cd_hit}/cluster_protein_90.fa \
      --outfmt 6 --threads 36 --max-target-seqs 1 --quiet -e 1e-5 --sensitive \
      --out ~/WORKSPACE/result/ff_xylan_pig/dbcan2/dbCAN_diamond


##提取基因list
cut -f 1,2 ~/WORKSPACE/result/ff_xylan_pig/dbcan2/dbCAN_diamond | uniq | sed 's/|/\t/g' | cut -f 1,3 | cut -f 1,2 -d ' ' | sed '1 i Name\tCAZ' > ~/WORKSPACE/result/ff_xylan_pig/dbcan2/gene_list

##合并基因list和基因丰度表
awk 'BEGIN{FS=OFS="\t"} NR==FNR{a[$1]=$2} NR>FNR{print $0,a[$1]}' ~/WORKSPACE/result/ff_xylan_pig/dbcan2/gene_list ~/WORKSPACE/result/ff_xylan_pig/salmon/result/gene.TPM \
| sed '/\t$/d' > ~/WORKSPACE/result/ff_xylan_pig/dbcan2/gene_list_TPM

##合并基因list和基因counts表
awk 'BEGIN{FS=OFS="\t"} NR==FNR{a[$1]=$2} NR>FNR{print $0,a[$1]}' ~/WORKSPACE/result/ff_xylan_pig/dbcan2/gene_list ~/WORKSPACE/result/ff_xylan_pig/salmon/result/gene.count \
| sed '/\t$/d' > ~/WORKSPACE/result/ff_xylan_pig/dbcan2/gene_list_count