Background

Maize is a highly polymorphic species with numerous single nucleotide polymorphisms (SNP) and insertion/deletions (InDel) commonly present among maize alleles [1, 2]. One consequence of this level of polymorphism is that sequence alignments of maize genes constructed with standard software such as CLUSTAL W, Muscle, Dialign, MAFFT, T-Coffee, etc., must be visually inspected and manually edited to ensure that an optimal alignment is obtained. This problem, together with the need to make base calls and alignment decisions in regions of variable base quality, creates the need for a tool to assist the researcher in manual editing of sequence alignments. Researchers in communities other than maize also require user-friendly tools to facilitate input, processing, and output of the sequence diversity among individuals within the species. Although there are several improved methods [3] to generate alignments for multiple nucleotide sequences, there is a need to improve software for manual editing to evaluate and fix discrepancies. Our goal was to produce a sequence alignment tool (DNAAlignEditor) that provides an efficient path for editing multiple sequence alignments to improve alignment quality, tracking edited alignments, and exporting sequence files to external tools for analysis.

Implementation

DNAAlignEditor is written in Visual Basic for the user interface, MS Access database for the stand-alone version or PostrgeSQL 8.0 database for the network version, and embedded ActiveX data object drivers for connectivity between the database and the user interface. DNAAlignEditor network version is designed to share alignment information when several users are simultaneously logged into the system and to manage alignments collected in the database. Regardless of whether the stand-alone or network version is used, the database aids in the tracking and manipulation of large numbers of sequences, and the relationship among genes, primers, amplicons, and population groups. The database is also required to track multiple versions of the same alignments and to combine alignments. Editing of a handful of gene alignments may be accomplished without the aid of a database; however, processing, editing, and finishing hundreds of alignments requires a more sophisticated system managed by a database.

DNAAlignEditor stores a set of interrelated tables in a relational database (DNADB) supporting alignment information to provide easy editing, querying, importing, and exporting function from genes, primers, populations, nucleotide sequence alignments, and quality scores as shown in the entity relationship diagram [see additional file 1]. The system is a user-friendly editing tool using grids, text boxes, combo boxes, and scroll lists to facilitate data access. DNAAlignEditor tool color codes nucleotide identity for ease in visual identification of sequence polymorphisms and displays the quality score associated with each base to aid in manual editing. The quality score-centric view is one of the major and unique features that distinguish DNAAlignEditor from other alignment programs (Figure 1).

Figure 1
figure 1

DNAAlignEditor view of an example alignment. The color coded cells represent the base calls and the numbers within the cells reflect the quality scores. Insertion/deletions are present in columns 576–578 and 591, and SNPs in columns 579, 580, 594, 597, 599 and 603. Below the alignment are the buttons used to manipulate sequence. To the left of the alignment are the names of the individuals displayed. Above the alignment are identification summary statistics on the alignment and additional buttons to manipulate sequence file and re-CLUSTAL functions.

Full size image

DNAAlignEditor accepts alignments in three different formats and checks the consistency of the information before addition to the database [see additional file 2]. The first format uses the information from PHRAP ace files, and a file that defines the gene information associated with the alignment. The second format is a fasta format. The last format accepts a tab delimited text, file having a header with information related to gene, amplicon, and total bases followed by a nucleotide sequence row and a quality scores row (or no-quality where a default quality score of 15 is applied to all base calls) [see additional file 2].

The program validates the data before it is imported, i.e. that genes, amplicons, and populations exist in DNADB. For sequences without quality scores, the program assigns a default quality score of 15 for each base. This value can be changed in the configuration file. By definition, a phred quality score of Q ≥ 20 indicates that there is a less than 1 in 100 chance that the base call is incorrect. Consequently, "Q20" bases are considered "high quality" calls with corresponding low error probabilities [4, 5]. By adding a standard quality score of Q = 15, we effectively flag those bases as unknown quality.

Once the alignment information is in the database, it can be edited in an intuitive, user-friendly interface. A user help file can be accessed by pressing the F1 key. The tool provides several features to edit information stored in the database such as moving individual sequences up and down for ease of fixing errors in multiple lines, moving bases from the 5' direction to the 3' direction and conversely, deleting and inserting individual bases and block of bases, and deleting and inserting entire columns in the alignment. DNAAlignEditor forms initial alignments by executing either of the integrated CLUSTAL W [6] or MUSCLE programs [7], and displays and prints phylogenies with TreeView [8]. The inclusion of a "Re-CLUSTAL" function allows editing to be preformed as a reiterative process, for instance rerunning CLUSTAL W after trimming end sequences or removing lower quality individual sequences for the initial alignment. DNAAlignEditor also allows the user to export fasta files with quality files and/or files compatible with DnaSP [9] for further analysis. Upon editing changes and saving, the tool will generate a new alignment while maintaining the original for further use or reference.

Results and discussion

There are several software packages that produce nucleotide sequence alignments such as CLUSTAL W, Muscle, Dialign, MAFFT, and T-Coffee. However, all of these programs produce alignments that must be manually examined and edited for accuracy. By integrating a highly intuitive visual display with CLUSTAL functions, DNAAlignEditor allows the researcher move seamlessly between alignment and editing functions. The inclusion of quality scores allow the researcher to make informed decisions on editing InDel containing regions and accessing the certainty of unique nucleotide polymorphisms. The database functions promote efficient tracking of large sets of primers, genes, amplicons; and the relationships among these components of diversity sequencing projects. It is the focus on intuitive manual editing functions, inclusion of quality scores, and power of tracking components in large project that distinguish the software from other available for alignment functions. While a variety of alignment editors are now available include BioEdit [10], JalView [11], and INTERALIGN [12]. Of these only JalView, like DNAAlignEditor, is open source. None of the other programs is as intuitive to the naïve user nor uses the quality score feature central to DNAAlignEditor. In addition, the database functions of DNAAlignEditor for tracking the components of a re-sequencing projects makes DNAAlignEditor more appropriate for large DNA sequence diversity projects.

This software has been successfully used by the National Science Foundation funded "Molecular and Functional Diversity of the Maize Genome" group at the University of Missouri [13], University of Wisconsin, and Cold Spring Harbor Laboratories for manual editing of alignments. DNAAlignEditor was essential for tracking and exporting over 4,000 alignments are currently stored in DNADB with more than 50,000 sequences from several maize, teosinte, and Tripsacum accessions. In addition the amplicon merge feature was very helpful for extended sequencing of genes of particular interest to the group. The finished alignments are displayed in a comprehensive graphical user interface tool included in panzea site [14].

Conclusion

We have developed a nucleotide sequence alignment editor (DNAAlignEditor) that provides an intuitive, user-friendly tool for manual editing of multiple sequence alignments which can be used for different species to allow researchers to visualize and edit the alignment information in a straight forward way, and realign the sequences using the CLUSTAL W or Muscle software, as well as export information to different packages for further analysis.

Availability and requirements

Project name: DNAAlignEditor executable and source code are freely available at http://maize.agron.missouri.edu/~hsanchez/DNAAlignment_Tool.html [See additional files 3, 4]

Contact: sanchezvilledah@missouri.edu

Operating systems: Windows 95/98/NT/2000/XP; can run on Intel Mac using Parallels

Programming language: Visual Basic 6.0