Draft:SimPlot++

SimPlot++. is a free multi-platform application which allows the users to generate publication-quality sequence similarity plots. This open-source program was developed at the department of Computer Science of the Université du Québec à Montréal. SimPlot++ incudes 20 amino acid and 63 nucleotide evolutionary models. It allows the users to draw high-quality sequence similarity plots, identify recombination events (χ2, Neighbour Similarity Score , Pairwise Homoplasy Index and Proportion recombinations tests are available), and visualize and investigate sequence similarity networks. Improved versions of the SimPlot, FindSites and BootScan tools (originally proposed in the SimPlot software) have been implemented and made available in SimPlot++.

The Windows, Mac OS and Linux versions of the program can be freely downloaded from github. SimPlot++ is a new addition to other similarity plot software tools, including RIP, SimPlot , BLAST Genotyping tool , T-RECs and RDP5 , which have been widely used in many bioinformatics and evolutionary studies.

General Use
Simplot++ takes as input a multiple sequence alignment of DNA or Amino acid sequences. The input sequences should be aligned and encoded using one of the following data formats: FASTA, Clustal, Nexus, Stockholm, PIR or PHYLIP.

The input sequences loaded into SimPlot++ should be separated in different groups by the user, based on their evolutionary resemblance. This will allow the program to create consensus sequences. (i.e. group sequences). The consensus sequences will then be used to carry out all analyses available in SimPlot++. The minimum number of groups which should be created is two (to get full access to most of the program features)

SimPlot analysis


The multiple sequence alignment (MSA) (MSA) provided by the user is processed by means of a sliding window-based algorithm using as parameters a specified progress step and a sliding window size to carry out SimPlot analysis. The portion of the MSA corresponding to the current location of the sliding window is used to compute an evolutionary distance matrix with respect to the selected evolutionary distance model. This matrix is processed by the program to generate a sequence similarity plot that shows similarity between all available individual and consensus sequences and the selected reference sequence.

To conduct the SimPlot analysis one can use 63 DNA and 20 amino acid evolutionary models, activate a multiprocessing option to speed up the calculation process, or take advantage of Matplotlib-based sequence similarity graphs which can be easily modified and saved in a variety of formats. SimPlot++ also offers a novel distance calculability diagnostic option to show for which sequence regions the distance calculation was impossible (if any).

BootScan analysis


The BootScan analysis available in SimPlot++ relies on the four following steps. The consensus sequence fragments covered by the current position of the sliding window (i.e. the sub-MSA corresponding to the current position of the sliding window) are bootstrapped n times. For each such a sub-MSA, the corresponding distance matrix is generated and a phylogeny is constructed using either Neighbor joining or UPGMA algorithms. The number of trees in which the reference sequence is the nearest neighbor of each consensus sequence is calculated. The same variety of evolutionary distance models, the multiprocessing capability, and the similarity plot options as for SimPlot analysis are also available with Bootscan analysis. Importantly, the selected sequences must be aligned before carrying out the BootScan analysis.

FindSites
The FindSites analysis is usually carried out to detect eventual recombined sequence regions by using the Informative sites method. Four sequences of interest are involved in the analysis. One of them is supposed to be a sequence resulting from recombination, two of them are supposed to be the parents of the first sequence, and the forth one plays the role of an outgroup. The Informative sites method identifies the sites of interest as those at which two of the four sequences share the same nucleotides, while the other two share a different one.

Similarity Network


Sequence similarity networks can be used to represent similarity relationships by visualizing different similarity links between evolutionary sequences. In a sequence similarity network every sequence, or group of sequences, is represented by a single vertex (or network node). A given pair of vertices in such a network can be linked (or not linked) to each other by a branch depending on the value of the evolutionary distance between these sequences and the similarity threshold selected by the user. Both local (when only some sequence regions are considered) and global sequence similarity can be represented using this kind of graphs. The similarity threshold can be changed to better visualize either local or global similarity connections. Moreover, a similarity network can correspond to a selected sequence region of a given MSA. This can help one analyze in greater details a specific gene or gene region of a given MSA. The networks produced by SimPlot++ can be saved in the png, svg or HTML formats. Importantly, the selected sequences must be aligned before carrying out the network analysis. It is worth noting that the sequence similarity network analysis is available in SimPlot++ only (it is not available in SimPlot ).

Recombination analysis
Several statistical tests for identifying recombination (i.e. mosaic sequence regions ) have been included in SimPlot++. Some of them come from the PhiPack package. These tests include the Phi (Pairwise Homoplasy Index) test, the Phi-profile (Pairwise Homoplasy Index-profile) test, the Neighbour Similarity Score (NSS) test and the Maximum χ2 test. These tests can be conducted for both ungrouped (i.e. original) and grouped sequences. Importantly, the selected sequences must be aligned before carrying out the recombination analysis. In addition, a novel fast Proportion test can be also carried out to detect recombination events in large genomic sequences. Once again, the described options of recombination analysis are available in SimPlot++ only (they are not available in SimPlot ). Other existing programs for detecting recombination include: TOPALi, RAT , T-RECs and RDP5