根据蛋白多序列比对文件mapping颜色至蛋白三维结构

杨小小

省略一千字。

使用场景

蛋白氨基酸多序列比对能体现不同蛋白间的进化关系，以及用于发现蛋白结构中的保守结构和容易发生突变的结构。

蛋白氨基酸多序列比对

首先从 UniProt 网站上搜索需要进行比对的蛋白序列。

2. 蛋白多序列比对的软件和网络服务器有很多，可以自行百度。这里使用 Clustal Omega 进行蛋白多序列比对。将输出的结果文件保存为 .fas 格式文件。如下图所示：

Mapping多序列比对文件至蛋白三维结构

(1) 从 RCSB晶体结构库中下载最完整，分辨率最高的蛋白晶体结构 (如没有结构的可用 AlphaFold2 进行预测)，在 Pymol 中去溶剂，去小分子，和uniprot序列肉眼比对保存单链为.pdb。

(2) 使用 ConSurf 网站将多序列比对信息mapping到蛋白三维结构中。具体操作如下图所示：

输入：

输出：

获取高分辨率图片

(1) 下载输出文件，解压，其中.html中是多序列比对的结果，_ATOMS_section_With_ConSurf.pdb中含有mapping的颜色的信息。

(2) 将_ATOMS_section_With_ConSurf.pdb加载进Pymol并运行下述脚本consurf_new.py (run consurf_new.py)。

# consurf_new.py 
# Define a Python subroutine to colour atoms by B-factor, using predefined intervals
def colour_consurf(selection="all"):
    # Colour other chains gray, while maintaining
    # oxygen in red, nitrogen in blue and hydrogen in white    
    cmd.color("gray", selection)
    cmd.util.cnc()
    # These are constants
    minimum = 0.0
    maximum = 9.0
    n_colours = 9
    # Colours are calculated by dividing the RGB colours by 255
    # RGB = [[16,200,209],[140,255,255],[215,255,255],[234,255,255],[255,255,255],
    #        [252,237,244],[250,201,222],[240,125,171],[160,37,96]]
    colours = [
                [0.039215686, 0.490196078, 0.509803922],
                [0.294117647, 0.68627451, 0.745098039],
                [0.647058824, 0.862745098, 0.901960784],
                [0.843137255, 0.941176471, 0.941176471],
                [1, 1, 1],
                [0.980392157, 0.921568627, 0.960784314],
                [0.980392157, 0.784313725, 0.862745098],
                [0.941176471, 0.490196078, 0.666666667],
                [0.62745098, 0.156862745, 0.37254902]]
    bin_size = (maximum - minimum) / n_colours
    # Loop through colour intervals
    for i in range(n_colours):
        lower = minimum + (i + 1) * bin_size
        upper = lower + bin_size
        colour = colours[i]
        # Print out B-factor limits and the colour for this group
        print(lower, " - ", upper, " = ", colour)
        # Define a unique name for the atoms which fall into this group
        group = selection + "_group_" + str(i + 1)
        # Compose a selection command which will select all atoms which are
        #	a) in the original selection, AND
        #	b) have B factor in range lower <= b < upper
        sel_string = selection + " & ! b < " + str(lower)
        if(i < n_colours):
            sel_string += " & b < " + str(upper)
        else:
            sel_string += " & ! b > " + str(upper)
        # Select the atoms
        cmd.select(group, sel_string)
        # Create a new colour
        colour_name = "colour_" + str(i + 1)
        cmd.set_color(colour_name, colour)
        # Colour them
        cmd.color(colour_name, group)
    # Create new colour for insufficient sequences
    # RGB_colour = [255,255,150]
    insuf_colour = [1, 1, 0.588235294]
    cmd.set_color("insufficient_colour", insuf_colour)
    # Colour atoms with B-factor of 10 using the new colour
    cmd.select("insufficient", selection + " & b = 10")
    cmd.color("insufficient_colour", "insufficient")