Home 生物信息

Subtitle: Predict Secretory Protein in microbes with SignalP and TMHMM

分泌蛋白Secretory Protein是指在细胞内合成后，分泌到细胞外起作用的蛋白质。分泌蛋白的 N 端有一般由 15～30 个氨基酸组成的信号肽。信号肽是引导新合成的蛋白质向分泌通路转移的短（长度 5-30 个氨基酸）肽链。常指新合成多肽链中用于指导蛋白质的跨膜转移（定位）的 N - 末端的氨基酸序列（有时不一定在 N 端）。使用 SignalP 注释蛋白序列是否含有信号肽结构，使用 TMHMM 注释蛋白序列是否含有跨膜结构，最终筛选出含有信号肽结构并且不含跨膜结构的蛋白为分泌蛋白。

#软件 Software

  • SignalP V6.0

  • SignalP 6.0预测来自古细菌、革兰氏阳性细菌、革兰氏阴性细菌和真核生物的蛋白质中存在的信号肽predicts signal peptides and the location of their cleavage sites in proteins from Archaea, Gram-positive Bacteria,及其切割位点的位置。Gram-negative Bacteria and Eukarya.在细菌和古细菌中，SignalP 6.0 可以区分五种类型的信号肽：In Bacteria and Archaea, SignalP 6.0 can discriminate between five types of signal peptides:

    • Sec/SPI：由 Sec 转座转运，并由信号肽酶 I (Lep) 切割的 “标准” 分泌信号肽；"Standard" secretory signal peptides transported by Sec translocon and cleaved by Signal Peptidase I (Lep).

    • Sec/SPII：由 Sec 转座子运输，并由信号肽酶 II (Lsp) 切割的脂蛋白信号肽；lipoprotein signal peptides transported by the Sec translocon and cleaved by Signal Peptidase II (Lsp).

    • Tat/SPI：由 Tat 转座子转运，并由信号肽酶 I (Lep) 切割的 Tat 信号肽；Tat signal peptides transported by the Tat translocon and cleaved by Signal Peptidase I (Lep).

    • Tat/SPII：由 Tat 转位子转运，并由信号肽酶 II (Lsp) 切割的 Tat 脂蛋白信号肽；Tat lipoprotein signal peptides transported by Tat translocon & cleaved by Signal Peptidase II (Lsp).

    • Sec/SPIII：由 Sec 转位子运输，并由信号肽酶 III (PilD/PibD) 切割的菌毛蛋白和菌毛蛋白样信号肽。Pilin & pilin-like signal peptides transported by Sec translocon & cleaved by Signal Peptidase III (PilD/PibD).

    • 此外，SignalP 6.0 预测信号肽的区域。Additionally, SignalP 6.0 predicts the regions of signal peptides.根据类型，预测 n、h 和 c 区域以及其他显着特征的位置。Depending on the type, the positions of n-, h- and c-regions as well as of other distinctive features are predicted.

  • TMHMM V2.0c

    • 用于预测蛋白质中的跨膜螺旋。

  • Python

SignalP 和 TMHMM 对于学术用户免费，但是需要填写相关信息和邮箱，以接收下载链接（4h 有效时间）。

#软件安装 Installation of Softwares

# 安装 SignalP 6.0

  • 下载

    访问 SignalP V6.0 网站，找到 “Download”，填写相关信息，获取下载链接，下载得到 “signalp-6.0.fast.tar.gz”。有两个模式可以选择 ——“slow_sequential” 和 “fast"。前者 runs the full model sequentially, taking the same amount of RAM as fast but being 6 times slower；后者 uses a smaller model that approximates the performance of the full model, requiring a fraction of the resources and being significantly faste。本教程下载的是 fast 模式。

  • 安装 Installation

    • 安装依赖 Dependencies

      • Python

      • matplotlib>3.3.2

      • numpy>1.19.2

      • torch>1.7.0

        pip install torch

      • tqdm>4.46.1

    • 安装 SignalP 6.0

      # 解压缩安装文件
      tar zxvf signalp-6.0.fast.tar.gz
      # 进入解压后的软件目录，在终端运行
      python setup.py install
      # 测试安装
      signalp6 --help
# 安装 TMHMM V2.0c
  • 下载
    访问 TMHMM V2.0c 网站，找到 “Download”，填写相关信息，获取下载链接，下载得到 “tmhmm-2.0c.Linux.tar.gz”。
  • 安装
    # 解压缩
    tar zxvf tmhmm-2.0c.Linux.tar.gz
    # 进入解压后的目录
    cd tmhmm-2.0c
    # 获取当前路径，我的是 “/home/liu/tools/tmhmm-2.0c/bin”
    pwd
    # 将该路径加入到系统的环境变量中，参考我之前的文章来（编辑～/.bashrc）https://liaochenlanruo.github.io/post/f6c9.html#% E6% B7% BB% E5%8A% A0% E7%8E% AF% E5% A2%83% E5%8F%98% E9%87%8F
    # 修改 bin 目录下的 tmhmm 和 tmhmmformat.pl 的首行为 “#!/usr/bin/perl”
  • 运行错误
    运行软件时总报 Segmentation fault (core dumped) 错误，暂时无解。各位可以使用其在线版。
#软件用法 Usage
# SignalP 6.0
#预测 Prediction
A command takes the following form
signalp6 --fastafile /path/to/input.fasta --organism other --output_dir path/to/be/saved --format txt --mode fast
  • fastafile输入文件为 FASTA 格式的蛋白序列文件Specifies the fasta file with the sequences to be predicted.。
  • organism is either other or Eukarya . Specifying Eukarya triggers post-processing of the SP predictions to prevent spurious results (only predicts type Sec/SPI).
  • format can take the values txt , png , eps , all . It defines what output files are created for individual sequences. txt produces a tabular .gff file with the per-position predictions for each sequence. png , eps , all additionally produce probability plots in the requested format. For larger prediction jobs, plotting will slow down the processing speed significantly.
  • mode is either fast , slow or slow-sequential . Default is fast , which uses a smaller model that approximates the performance of the full model, requiring a fraction of the resources and being significantly faster. slow runs the full model in parallel, which requires more than 14GB of RAM to be available. slow-sequential runs the full model sequentially, taking the same amount of RAM as fast but being 6 times slower. If the specified model is not installed, SignalP will abort with an error.
#输出Outputs
  • output_dir/output.gff3：仅包含含有信号肽的序列信息；
    output.gff3
  • output_dir/prediction_results.txt：包含了输入文件中的所有序列（不重要）；
  • output_dir/region_output.gff3：包含所有的信号肽区域信息。
    • n-region: The n-terminal region of the signal peptide. Reported for Sec/SPI, Sec/SPII, Tat/SPI and Tat/SPII. Labeled as N
    • h-region: The center hydrophobic region of the signal peptide. Reported for Sec/SPI, Sec/SPII, Tat/SPI and Tat/SPII. Labeled as H
    • c-region: The c-terminal region of the signal peptide, reported for Sec/SPI and Tat/SPI.
    • Cysteine: The conserved cysteine in +1 of the cleavage site of Lipoproteins that is used for Lipidation. Labeled as c.
    • Twin-arginine motif: The twin-arginine motif at the end of the n-region that is characteristic for Tat signal peptides. Labeled as R.
    • Sec/SPIII: These signal peptides have no known region structure.
      region_output.gff3
# 批处理与结果优化
脚本名：run_SignalP.pl
#!/usr/bin/perl
use strict;
use warnings;
# Author: Liu Hualin
# Date: Oct 14, 2021
open IDNOSEQ, ">IDNOSEQ.txt" || die;
my @faa = glob("*.faa");
foreach  (@faa) {
	$_ =~ /(.+).faa/;
	my $str = $1;
	my $out = $1 . ".nodesc";
	my $sigseq = $1 . ".sigseq";
	my $outdir = $1 . "_signalp";
	open IN, $_ || die;
	open OUT, ">$out" || die;
	while (<IN>) {
		chomp;
		if (/^(>\S+)/) {
			print OUT $1 . "\n";
		}else {
			print OUT $_ . "\n";
		}
	}
	close IN;
	close OUT;
	my %hash = idseq($out);
	system("signalp6 --fastafile $out --organism other --output_dir $outdir --format txt --mode fast");
	my $gff = $outdir . "/output.gff3";
	if (! -z $gff) {
		open IN, "$gff" || die;
		<IN>;
		open OUT, ">$sigseq" || die;
		while (<IN>) {
			chomp;
			my @lines = split /\t/;
			if (exists $hash{$lines[0]}) {
				print OUT ">$lines[0]\n$hash{$lines[0]}\n";
			}else {
				print IDNOSEQ $str . "\t" . "$lines[0]\n";
			}
		}
		close IN;
		close OUT;
	}
	system("rm $out");
	system("mv $sigseq $outdir");
}
close IDNOSEQ;
sub idseq {
	my ($fasta) = @_;
	my %hash;
	local $/ = ">";
	open IN, $fasta || die;
	<IN>;
	while (<IN>) {
		chomp;
		my ($header, $seq) = split (/\n/, $_, 2);
		$header =~ /(\S+)/;
		my $id = $1;
		$hash{$id} = $seq;
	}
	close IN;
	return (%hash);
}
将 run_SignalP.pl 与后缀名为 “.faa” 的 FASTA 格式文件放在同一目录下，在终端中运行如下代码：
perl run_SignalP.pl
#结果解读Output interpretation
* 代表输入文件的名字。
  • *_signalp/output.gff3：仅包含含有信号肽的序列信息；
  • *_signalp/prediction_results.txt：包含了输入文件中的所有序列（不重要）；
  • *_signalp/region_output.gff3：包含所有的信号肽区域信息；
  • *_signalp/*.sigseq：存储所有信号肽的氨基酸序列文件，可用作 TMHMM 的输入文件。
# TMHMM
# 预测
离线版总是报错，找不出原因，因此使用网页服务器进行，输入文件为上述生成的 “*_signalp/*.sigseq”，将其上传至网页版 TMHMM，提交任务，等待结果即可。
# 结果展示
TMHMM 可以输出多种格式的结果文件，具体请参考其官方说明。
在TMHMM网站提交任务
  • Long output format
    • Length：蛋白序列的长度。The length of the protein sequence.
    • Number of predicted TMHs：预测到的跨膜螺旋的数量。The number of predicted transmembrane helices.
    • Exp number of AAs in TMHs：跨膜螺旋中氨基酸的预期数量。The expected number of amino acids intransmembrane helices.如果此数字大于 18，则很可能是跨膜蛋白（或具有信号肽）。If this number is larger than 18 it is very likely to be a transmembrane protein (OR have a signal peptide).
    • Exp number, first 60 AAs：在蛋白的前 60 个氨基酸中跨膜螺旋中氨基酸的预期数量。The expected number of amino acids in transmembrane helices in the first 60 amino acids of the protein.如果该数字超过几个，你应该被警告在 N 端预测的跨膜螺旋可能是一个信号肽。If it more than a few, you are warned that a predicted transmembrane helix in the N-term could be a signal peptide.
    • Total prob of N-in：N 端在膜的细胞质一侧的总概率。The total probability that the N-term is on the cytoplasmic side of the membrane.
    • POSSIBLE N-term signal sequence：当 “Exp number, first 60 AAs” 大于 10 时产生的警告。 A warning that is produced when "Exp number, first 60 AAs" is larger than 10.
  • 蛋白 F01_bin.1_00110 共计 436 个氨基酸，有 5 个跨膜螺旋结构。
    含有跨膜结构的蛋白
  • 蛋白 F01_bin.1_00142 共计 557 个氨基酸，所有序列均在膜外，即该序列编码的是分泌蛋白。
    不含跨膜结构的蛋白
  • Short output format
    • "len="：蛋白序列的长度。 The length of the protein sequence.
    • "ExpAA="：跨膜螺旋中氨基酸的预期数量。The expected number of amino acids intransmembrane helices.如果此数字大于 18，则很可能是跨膜蛋白（或具有信号肽）。If this number is larger than 18 it is very likely to be a transmembrane protein (OR have a signal peptide).
    • "First60="：在蛋白的前 60 个氨基酸中跨膜螺旋中氨基酸的预期数量。The expected number of amino acids in transmembrane helices in the first 60 amino acids of the protein.如果该数字超过几个，你应该被警告在 N 端预测的跨膜螺旋可能是一个信号肽。If it more than a few, you are warned that a predicted transmembrane helix in the N-term could be a signal peptide.
    • "PredHel="：预测到的跨膜螺旋的数量。 The number of predicted transmembrane helices by N-best.
    • "Topology="：N-best 预测的拓扑结构。The topology predicted by N-best.拓扑是由跨膜螺旋的位置给出的，如果螺旋在内部，则由 “i” 分隔，如果螺旋在外部，则由 “o” 分隔。'i7-29o44-66i87-109o' 意味着它从膜内开始，在位置 7 到 29 有一个预测的 TMH，30-43 在膜外，然后是位置 44-66 的 TMH。
    Short output format
# 结果汇总
通过网页版预测我们仅得到了一个列表文件（Short output format），该文件需要自己复制网页内容粘贴到新文件中，我将其命名为*_TMHMM_SHORT.txt，并将其存放在*_signalp目录中，该目录是由 run_SignalP.pl 生成的。下面我将会统计各个基因组中信号肽蛋白的总数量、分泌蛋白数量和跨膜蛋白数量到文件 Statistics.txt 中，并分别提取每个基因组的分泌蛋白序列到*_signalp/*.secretory.faa文件中，提取跨膜蛋白序列到*_signalp/*.membrane.faa文件中。该过程将通过 tmhmm_parser.pl 完成。
#!/usr/bin/perl
use strict;
use warnings;
# Author: Liu Hualin
# Date: Oct 15, 2021
open OUT, ">Statistics.txt" || die;
print OUT "Strain name\tSignal peptide numbers\tSecretory protein numbers\tMembrane protein numbers\n";
my @sig = glob("*_signalp");
foreach my $sig (@sig) {
	$sig=~/(.+)_signalp/;
	my $str = $1;
	my $tmhmm = $sig . "/$str" . "_TMHMM_SHORT.txt";
	my $fasta = $sig . "/$str" . ".sigseq";
	my $secretory = $str . ".secretory.faa";
	my $membrane = $str . ".membrane.faa";
	open SEC, ">$secretory" || die;
	open MEM, ">$membrane" || die;
	my $out = 0;
	my $on = 0;
	my %hash = idseq($fasta);
	open IN, $tmhmm || die;
	while (<IN>) {
		chomp;
		$_=~s/[\r\n]+//g;
#		print $_ . "\n";
		my @lines = split /\t/;
		if ($lines[5] eq "Topology=o") {
			$out++;
			print SEC ">$lines[0]\n$hash{$lines[0]}\n";
		}else {
			$on++;
			print MEM ">$lines[0]\n$hash{$lines[0]}\n";
		}
	}
	close IN;
	close SEC;
	close MEM;
	system("mv $secretory $membrane $sig");
	my $total = $out + $on;
	print OUT "$str\t$total\t$out\t$on\n";
}
close OUT;
sub idseq {
	my ($fasta) = @_;
	my %hash;
	local $/ = ">";
	open IN, $fasta || die;
	<IN>;
	while (<IN>) {
		chomp;
		my ($header, $seq) = split (/\n/, $_, 2);
		$header =~ /(\S+)/;
		my $id = $1;
		$hash{$id} = $seq;
	}
	close IN;
	return (%hash);
}
运行方法：将tmhmm_parser.pl放在*_signalp的上一级目录下，*_signalp目录中必须包含*_TMHMM_SHORT.txt文件和*.sigseq文件。在终端运行如下代码：
perl tmhmm_parser.pl
# 脚本获取
关注公众号 “生信之巅”，聊天窗口回复 “c886” 获取下载链接。
生信之巅微信公众号生信之巅小程序码
敬告：使用文中脚本请引用本文网址，请尊重本人的劳动成果，谢谢！Notice: When you use the scripts in this article, please cite the link of this webpage. Thank you!
# 参考
  • SignalP V6.0
  • TMHMM
   
 Edited on   Views  times
Give me a cup of [coffee]~(￣▽￣)~*
Hualin Liu WeChat Pay
WeChat Pay
Hualin Liu Alipay
Alipay