With exponentially increasing genomic sequences available in the public domain [1–5] comparative genomics demonstrates its power to help biologists identify novel conserved and functional regions in genomes [6–9]. Based on the comparison of cross-species genomic sequences, biologists can understand the evolutionary relationship of genomic regions among species, discover conserved regions between different genomes, such as yeast species genomes , metazoan genomes , vertebrate genomes , and mammalian genomes , discover regulatory motifs in the yeast  and human promoters  or identify potential conserved non-genic sequences (CNGs) .
However, genomic sequences can be megabase long and thus the traditional sequence alignment tools based on dynamic programming would not work efficiently due to their time and space complexities. To better tackle this problem, several tools for genomic sequence alignment have been proposed, such as pairwise sequence aligners like MUMmer , GS-Aligner , Avid  and LAGAN , and multiple sequence alignment (MSA) programs like T-COFFEE, MAFFT , MultiPipMaker , MULTIZ , MLAGAN , MAVID , and MUSCLE [26, 27]. These alignment tools, however, are heuristics based and do not provide any indication of how far they are from an optimal solution. The comparisons of alignment tools using a set of benchmarking sequences have also been conducted in recent years [28–30]. We found that the majority of these tools usually fail to generate consistent results especially in aligning divergent cross-species sequences. As a result, the more alignment tools there are available in the public domain, the more confusion it creates for users to decide which tool is most suitable to align their sequences.
Although the comparison results in [28–31] provide some evaluations of several popular alignment tools, the conclusions may not be directly applicable to users' sequences. Furthermore the user usually does not know for sure whether those poorly aligned regions produced by the alignment tools are indeed non-homologous or just due to inappropriate tools or scoring systems used. Consequently, if some homologous regions are unaligned, the estimated evolution distances of these sequences may be inaccurate and therefore the constructed phylogenetic trees may be incorrect. Facing this problem, the user may have to try different tools or scoring systems to evaluate the correctness and accuracy of alignment results in the initial stage of sequence analysis. On the other hand, new alignment tools are released continually. Users may want to compare these newly released tools with those that they are most familiar with. Thus, it is desirable and most useful to have a visualization system that provides a direct and efficient method and can assist users to cross compare and inspect alignment results obtained by different MSA tools especially at the initial stage of sequence analysis.
In recent years, a number of visualization tools have been released in the public domain. These tools can be roughly divided into two categories: integrated genome/sequence browser and individual alignment result visualization. In the former category, such as UCSC ENCODE project [32, 33], UCSC human genome browser , Ensembl , ECR Browser [36, 37], users can view alignment results mapped onto the sequenced genomes. Some of these browsers also provide registered users to submit alignment results and see the conservation regions between different genomes. In the latter category, the tools are developed to visualize individual alignment results. The VISTA-related tools are among the famous ones that have been developed for several years . mVISTA is a set of programs for comparing DNA sequences from two or more species up to megabases long and visualize these alignments with annotation information . rVISTA (regulatory Vista) combines database searches for transcription factor binding sites with a comparative sequence analysis [40, 41]. GenomeVISTA compares users' sequences with several whole genome assemblies [42, 43]. Phylo-VISTA analyzes alignments of multiple DNA sequences from different species while considering their phylogenetic relationships . In general, the VISTA family of tools provides users with a novel graphical user interface (GUI) to view alignment results from different viewpoints. In addition to the VISTA family, PipMaker [23, 45], and zPicture  are also popular visualization tools for sequence or genomes alignment results. All of these tools are web-based with friendly user interfaces, and allow users to easily visualize alignment results with annotations. However, these tools are limited solely to single alignment results. The capability of simultaneously comparing multiple results from different alignment tools or different parameters of a scoring system, such as changing match rewards or mismatch penalties, is notably lacking.
In this article, we present a versatile alignment visualization system, SinicView (Sequence-aligning INnovative and Interactive Comparison VIEWer), which enables users to efficiently compare and evaluate assorted alignment results obtained by different tools. SinicView for the present calculates similarity of the alignment outputs under a fixed window using the sum-of-pairs method and provides scoring profiles of each set of aligned sequences. Other scoring matrices, such as EMBOSS DNA scoring matrix  and YASS , are also provided in SinicView for users to select. Besides, users can also upload their preferable scoring matrices to calculate the scoring profile curves. Users can visually compare alignment results either in graphic scoring profiles or in plain text format of the aligned nucleotides. In addition, the information about alignment gaps and sequence annotations is also presented. The real-time juxtaposition of the visualization results from different MSA programs would bring more insights into the evaluation process. With SinicView, users can use their own sequences to survey and compare various multiple alignment tools and thus to unveil their merits (and shortcomings). Moreover, the cross-tools comparison can provide users more confidence in their final alignment results especially for those poorly aligned regions.