Source of content: Guangdong Pharmaceutical University Laboratory Instruction Manual.
Operation method
Multiple sequence alignment and phylogenetic tree construction experiments
Principle
In modern molecular evolutionary studies, reconstructing the evolutionary history of organisms based on the genetic or species diversity of existing organisms is a very important issue. A reliable inference of phylogeny will reveal the sequence of the evolutionary processes of the organisms in question and help us to understand the history of biological evolution and evolutionary mechanisms. Move I. Multiple sequence comparison of known DNA sequences was done using CLUSTALX software. Common Problems I. For a complete evolutionary tree analysis the following steps are required (1) The multiple sequence targets to be analyzed should be compared (alignment). For more product details, please visit Aladdin Scientific website.
Operation steps:
1. Prepare 8 DNA sequence test.seq (or txt) files in FASTA format. 
2. Double-click to enter the CLUSTALX program, click FILE to enter LOAD SEQUENCE, and open the test.seq (or txt) file. 
3. Click ALIGNMENT, under Default Alignment Parameters, click Do complete Alignment. In the new window, click ALIGN to compare, this time the output of two files (the default output file format for Clustal format): comparison file test.aln and the wizard tree file test.dnd. 
4. Click FILE to enter Save sequence as, select PHYLIP in the format box, and the file will exist as test.phy in the PHYLIP software directory, click OK.
5. Copy the test.phy file from the PHYLIP software directory to the EXE folder. The partial sequence of the test.phy file opened with a notepad is as follows: 
The 8 and 50 in the figure indicate 8 sequences and each sequence has 50 bases respectively.
Second, use PHYLIP software to derive the evolutionary tree.
1. Enter the EXE folder, click the SEQBOOT program to enter the name of the test.phy file, and enter. 
The D, J, R, I, O, 1, 2 in the figure represent the selectable options, type these letters, the program conditions will change. option D need not be changed. option J has three conditions to choose from, respectively, Bootstrap, Jackknife and Permute. the article mentioned above the Bootstraping method to evaluate the evolutionary tree, the so-called Bootstraping method is to choose any half of the bases (amino acids) from the whole sequence, and the remaining half of the sequence is randomly complemented to form a new sequence. In this way, one sequence can become many sequences. One multi-sequence group can also become many multi-sequence groups. According to some algorithm (Maximum Parsimony, Maximum Likelihood, Divided Pairing, or Neighbor Joining) an evolutionary tree can be generated for each multi-sequence group. By comparing the generated evolutionary trees, we get the most "realistic" evolutionary tree according to the majority-rule.Jackknife is another method for randomly selecting sequences. It differs from Bootstrap in that it does not fill in the remaining half of the sequence, but only generates a new sequence that is shortened by half. permute is another sampling method that serves a different purpose than Bootstrap and Jackknife, and will not be described here. the R option lets the user enter the number of republicates. The so-called republicate is a multisequence group generated using the Bootstrap method. Depending on the number of sequences contained in the multiple sequences, you can select a different republicate, here choose 200, enter Y to confirm the parameter and enter an odd number (such as 3) under Random number seed (must be odd). When we set up the conditions and press enter, the program starts to run and generates a file outfile in the EXE folder, which is opened with Notepad as follows: 
This file contains 200 republicates.
2. the file outfile is changed to infile. click the DNADIST program. Option M is to enter the number of republicates you just set. enter D to select data sets and enter 200. 
After setting the conditions, enter Y to confirm the parameters. The program starts running and generates an outfile in the EXE folder, part of which is as follows: 
Change the name of the outfile file to infile. To avoid duplication with the original infile file, change the original file name to infile1.
3. In the EXE folder, select the algorithm for inferring evolutionary trees by distance matrices and click on the NEIGHBOR program. Enter M to change the parameter, D to select data sets. enter 200. enter the odd seed 3. 
Enter Y to confirm the parameters. The program starts to run and produces two outputs, outfile and outtree, in the EXE folder. outtree file is a tree file, which can be opened with software such as treeview. outfile is an output report of the analysis results, which includes the tree and some other analysis reports, and can be opened directly with Notepad. Part of the content is as follows: 
4. change the name of the outfile file in the EXE folder to outfile1 to avoid being overwritten by the newly generated outfile file. Click the CONSENSE program. Enter Y to confirm the setting. outfile and outtree are newly generated in the EXE folder. the outfile file is opened in Notepad with the following contents: 
5. change the intree file name in the EXE folder to intree1, and change the outtree to intree. click the DRAWTREE program and enter the font1 file name as a parameter. Enter Y to confirm the parameter. The program starts running and the Tree Preview diagram appears. 
6. Click the DRAWGRAM program and enter the name of the font1 file as the parameter. Enter Y to confirm the parameter. The program starts running and the Tree Preview appears.
(2) To construct a phyligenetic tree.
Algorithms for constructing evolutionary trees are mainly divided into two categories: discrete character methods and distance methods.
The discrete character method means that the topological shape of the evolution tree is determined by the state of each base/amino acid in the sequence (e.g., a sequence may contain many enzymatic sites, and the existence or otherwise of each enzymatic site is determined by the state of several bases, that is, the state of a sequence base determines the state of its enzymatic site, when multiple sequences are analyzed, the topological shape of the evolution tree is also determined by the state of these bases). (the topological shape of the evolutionary tree when multiple sequences are analyzed is also determined by the state of these bases).
The distance-dependent approach means that the topology of the tree is determined by the evolutionary distance between two sequences. The length of the branches of an evolutionary tree represents the evolutionary distance.
Independent element methods include Maximum Parsimony methods and Maximum Likelihood methods; distance-dependent methods include the Unweighted Pairing Method (UPGMAM) and Neighbor-joining.
(3) To evaluate the evolutionary tree, Bootstraping method is mainly used.
The construction of evolutionary tree is a statistical problem, and the evolutionary tree we constructed is only an evaluation or simulation of the real evolutionary relationship. If we adopt an appropriate method, the constructed evolutionary tree will be close to the real "evolutionary tree".
Simulated evolutionary trees require a mathematical method to evaluate them. Different algorithms have different objectives.
Generally speaking, the maximum parsimony method is suitable for multiple sequences that meet the following conditions: i the bases of the sequences to be compared are not very different, ii there is an approximately equal mutation rate for each base in the sequence, iii there is no tendency for too many substitutions/transformations, and iv the sequences to be examined have a large number of bases (more than a few thousand); analyzing the sequences by the maximum likelihood method does not require any of the above, but it is extremely time consuming. This method is extremely time-consuming to compute. If a large number of sequences are analyzed, it may take several days to complete the calculations.
UPGMAM (Unweighted pair group method with arithmetic mean) assumes that all nucleotides/amino acids have the same mutation rate during evolution, i.e. there is a molecular clock. The evolutionary tree obtained by this algorithm is relatively not very accurate and is now rarely used. Neighbor-joining is a frequently used algorithm that constructs relatively accurate evolutionary trees and is quick to compute. Its disadvantage is that all sites on the sequence are treated equally and, moreover, the evolutionary distance of the analyzed sequence cannot be too large. In addition, it should be noted that no existing algorithm may be well suited to it for some specific multiple sequence objects.
The above tree-building steps can be realized with CLUSTALX and PHYLIP software, which is a multiple sequence comparison software for Windows.
PHYLIP is a compressed package of several software packages, which is extremely powerful and consists of five main areas of functional software: i) Software for analyzing DNA and protein sequence data. ii) Software for analyzing distance data after the sequence data has been converted to distance data. iii) Software for analyzing gene frequencies and contiguous elements. iv) Software for viewing each base/amino acid of the sequence independently (base/amino acid states of 0 and 1 only) when analyzing the sequence. v, Software for analyzing the sequence according to the DOLLO parsimony algorithm. vi, Software for drawing and modifying the evolutionary tree.
II. Assignments
1. the results of phylogenetic tree construction using the DNA sequences given in the example above. (Include the results of sequence comparison and the final tree generated)
2. the protein sequences given below are used to construct a phylogenetic tree using the method above. (Including the sequence comparison results and the final generated tree)
>RAT
MEPKRIREGYLVKKGSVFNTWKPMWVVLLEDGIEFYKKKSDNNPKGMIPLKGSTLTSPCQDFGKRMFVLK
ITTTKQQDHFFQAAYLEERDAWVRDIKKAIKCIEGGQKFARKSTRRSIRLPETIDLGALYLSMKDPEKGI
>HUMAN
MEPKRIREGYLVKKGSVFNTWKPMWVVLLEDGIEFYKKKSDNSPKGMIPLKGSTLTSPCQDFGKRMFVFK
ITTTKQQDHFFQAAFLEERDAWVRDIKKKAIKCIEGGQKFARKSTRRSIRLPETIDLGALYLSMKDTEKGI
>CANFA
MEPKRIREGYLVKRGSVFNTWKPMWVVLLEDGIEFYKKKSDNSPKGMIPLKGSTLTSPCQDFGKRMFVFK
ITTTKQQDHFFQAAFLEERDSWVRDTKKKAIKCIEGGQKFARKSTRRSIRLPETVDLGALYLSMKDIEKGI
>MOUSE
MEPKRIREGYLVKKGSVFNTWKPMWVVLLEDGIEFYKKKSDNSPKGMIPLKGSTLTSPCQDFGKRMFVLK
ITTTKQQDHFFQAAFLEERDAWVRDIKKKAIKCIEGGQKFARKSTRRSIRLPETIDLGALYLSMKDPEKGI
>Canis
MEPKRIREGYLVKRGSVFNTWKPMWVVLLEDGIEFYKKKSDNSPKGMIPLKGSTLTSPCQDFGKRMFVFK
ITTTKQQDHFFQAAFLEERDSWVRDTKKKAIKCIEGGQKFARKSTRRSIRLPETVDLGALYLSMKDIEKGI
>Gallus gallus
MEREPMRIREGYLVKKGSMFNTWKPMWVVLLEDGIEFYKRKSDNSPKGMIPLKGSTINSPCQDFGKRMFV
FKLTAAKQQDHFFQASYLEERDAWVRDIKKAIQCIDGGQRFARKSTRKSIRLPETINLSALYLSMKDPEK
>Danio rerio
MEPTTIREGYLVKKGTVLNSWKAVWVVLKDDAIEFFKKKTDRNAKGMIPLKGATLTSPCQDFSKRALVFK
VSTAKNQDHYFQATHLEHEREHWVKDIRRAITCLQGGKKFARKSTRRSIRLPESVNLSELYVCMKDPDRGV
>chimpanzee
MEPKRIREGYLVKRGSVFNTWKPMWVVLLEDGIEFYKKKSDNSPKGMIPLKGSTLTSPCQDFGKRMFVFK
ITTTKQQDHFFQAAFLEERDAWVRDMKKAIKCIEGGQKFARKSTRRSIRLPETIDLGALYLSMKDTEKGI
3. The above method of constructing a phylogenetic tree is the N-J method. Summarize the differences between the procedures chosen for constructing a phylogenetic tree using protein sequences and those used for constructing a phylogenetic tree using DNA sequences.