ls>list.txt
# 编辑一下list.txt，保存需要处理的文件夹
list=$(cat list.txt)
# 都是clean data了，还是洗一下吧
for i in $list;do
  cd ${i};
  fastp -i ${i}_1.clean.fq.gz -o ${i}_R1.fq.gz -I ${i}_2.clean.fq.gz -O ${i}_R2.fq.gz -w 16;
  cd ..;
done

# 开始组装，
for i in $list;do
    cd ${i};
    time spades -o spades_result -1 ${i}_R1.fq.gz -2 ${i}_R2.fq.gz -t 20;
    cd ..;
done

# 开始busco，
conda activate buscoEnv
for i in $list;do
    cd ${i};
    time busco -i ./spades_result/contigs.fasta -o busco -m geno -c 20 -l ~/database/BUSCO/agaricales_odb10;
    cd ..;
done

# 统计每个基因在多少个物种中存在
dos2unix list2.txt
list=$(cat list.txt)
list2=$(cat list2.txt)
for i in $list2;do
    echo -ne "${i}\t" >> test.txt;
    find ./*/busco/run_agaricales_odb10/busco_sequences/single_copy_busco_sequences -name "${i}.faa"| wc -l >> genespe.txt;
done

# 把所有物种的每个基因单独放入一个文件中
dos2unix list3.txt
list3=$(cat list3.txt)
for i in $list3;do
    for j in $list;do
        echo ">${j}" >> ../all/${i}.fasta;
        grep -v ">" ../${j}/busco/run_agaricales_odb10/busco_sequences/single_copy_busco_sequences/${i}.faa >> ../all/${i}.fasta;
    done
done

# 序列比对
for i in *.fasta;do
    linsi --thread 20 ${i} > ${i}.mafft;
done


# 提取保守序列
for i in *.mafft;do
    Gblocks $i -b4=2 -b5=h -t=p;
done
# 序列合并
for i in *.mafft-gb;do
    seqkit sort $i > $i.sort;
done

for i in *.sort;do
    seqkit seq $i -w 0 > $i.seqkit;
done

paste -d " " *.seqkit > all.fa

# 建树！
# 先跑个模型
cd
java -jar prottest-3.4-2/prottest-3.4.2.jar -i  /mnt/d/Amanita/all/all.fa -all-distributions -F -AIC -BIC -tc 0.5 -threads 20 -o  /mnt/d/Amanita/all/prottest.out
# BIC选择的最佳模型JTT+I+G+F，就还是这个
raxmlHPC-PTHREADS-SSE3 -T 20 -f a -x 123 -p 123 -N 1000 -m PROTGAMMAIJTTF -k -O -n all.tre -s all.fa

# iqtree
# 测一下模型
iqtree -s all.fa -m TEST -nt 20
# 选择的结果模型为：LG+F+I+G4
iqtree -s all.fa -m LG+F+I+G4 -bb 1000 -nt 20
# 或者直接自动选择模型
iqtree -s all.fa -m MFP -bb 1000 -nt 20