Overview

Vector NTI Suite is a collection of four programs for the analysis, manipulation, design, and construction of DNA and protein molecules. Most of the functionality of this suite is found in the Vector NTI program. The other, smaller modules are Bioplot, AlignX, and ContigExpress. Bioplot allows visualization of physicochemical properties along polypeptide or DNA chains. AlignX provides for the alignment of multiple sequences using the ClustalW algorithm. ContigExpress (which is not available for the Macintosh) is used to assemble small DNA fragments obtained during a large sequencing project.

Vector NTI uses a database-oriented approach to sequence analysis. Instead of sequence and vector files being loaded and analyzed individually - the usual approach for this class of programs - all data is contained in a master database that can be browsed using the Database Explorer window. The master database contains tables of DNA/RNA molecules, protein molecules, enzymes, oligonucleotides, and gel markers. Within each table, you can create custom databases to store your favorite molecules or to organize your laboratory projects. Each database file can be opened in a tripartite window that simultaneously displays sequence data, structural data, and a graphical representation of the molecule.

The database approach of Vector NTI combined with the well-designed display window gives it some tremendous advantages over traditional file-based packages. Molecules can be created and modified from preexisting files in a matter of minutes by simply clicking and dragging molecular structures. For the design, construction, and tracking of modified vectors and molecules, Vector NTI has no peer. However, the program also lacks some basic functions, is very complex, and lacks adequate documentation. These characteristics, combined with its relatively high price, makes it less than an ideal choice as a general-purpose sequence analysis program.

How Long Did It Take to Learn to Use It Productively?

Approximately 1 week.

Product Quality

Customizability

Excellent. Because of its database-oriented design, the program evolves over time to contain all the molecules, probes, and enzymes that you use in your work. The idea is to have a single depository for all of your molecular data. In this respect, Vector NTI works very well and rapidly becomes a customized extension of an investigator's laboratory notebook.

Ability to Program in Scripts, Add Extension Modules, etc.

Fair. The program supports external "tools," a script-based mechanism to exchange information between other programs, such as email and Web browsers. The documentation does not explain how to create or modify these tools. The few scripts that are available on the company Web site or included with the program are rather specialized. There is no mechanism to either create scripts or modify the program's functionality.

Ability to Import and Export in Different File Formats

Very good. Vector NTI allows you to open and export files in the GenBank, GenPept, EMBL, SWISS-PROT, and FASTA formats. The program provides clipboard support for the same formats and also supports drag-and-drop transfer of files between suite modules and to the desktop. However, you cannot import files created by other molecular biology programs, and only the ContigExpress module can read ABI files. You can publish molecule files as Vector NTI Molecule Documents which can be read by a public-domain viewer or opened by a variety of different programs. Graphics can be exported by either the clipboard or as files.

Useful or Unusual Features

The first thing a user notices upon opening Vector NTI is the Database Explorer window. Instead of presenting you with a sequence window or a file dialog box, the program opens a window containing all of your data files. These files are organized into tables, which are specialized subbases listing DNA/RNA molecules, proteins, oligonucleotides, restriction enzymes, and markers for nucleic acid gels. Within each table is a collection of molecules that can be sorted, organized into custom subbases, documented, and manipulated. InforMax supplies Vector NTI with a useful collection of the most common commercial and public-domain vectors and restriction enzymes built in. You can also import a variety of different file types from the Internet or from other molecular biology programs.

Each file in the database is opened in a tripartite display window that simultaneously displays sequence information, structural details, and a graphical representation of the molecule in separate panes. The design of this display window is wonderful - easily the best aspect of Vector NTI's interface. If the imported file has structural annotations (or if these are added manually), the structural features are shown graphically. Clicking on the structure automatically selects the corresponding sequence in the appropriate pane. When a pane is active, a series of tool buttons become active in the toolbar above the window. These buttons provide access to most of the common operations specific to that pane. The buttons are small and mostly cryptically named, but fortunately hints pop up when you pause the cursor over any button. Because there is no visual indication to suggest which pane is active, the pattern of buttons is actually the best guide to what is going on in the program.

The Database Explorer function in Vector NTI allows you to sort, arrange, and manage all of your molecules. Enzymes, proteins, and nucleic acids can be imported and exported using a variety of different formats. You can add comments, user fields, and keywords to the individual database files, and search and arrange the files based on any combination of criteria. You can also find files based on molecular features and track parent-descendant relationships. Vector NTI also manages parent-descendant relationships, allowing you the option of propagating alterations through a collection of related molecules. Collection of files (subbases) can be archived and exchanged with other Vector NTI users. The program also supports the export of files as "Molecule Documents," an enhanced GenBank/GenPept format. Sequence information in these documents can be opened by any program that supports the GenBank or GenPept formats. Sequence and graphic information in the files can be opened using either Vector NTI or the public-domain Vector NTI Viewer. This makes "publishing" of molecule documents extremely simple. Overall, the data management aspects of Vector NTI are excellent.

Vector NTI was obviously designed to facilitate the construction and modification of DNA vectors. In this task it really shines. To build a new vector from a collection of preexisting DNA molecules, you simply click, copy, and paste. The program has a powerful built-in feature for designing new molecules. You specify what you want to create in the Goal Molecule Definition list. Help is provided by the Fragment Wizard, which prompts you for the appropriate information. You can either specify fragments to use or define structural features that you want to include. You can also specify what types of techniques you prefer to use (or those you do not want to use, such as partial digestion). The program then creates a construction strategy, which includes a list of procedures and various diagnostic options that you can use along the way - for instance, it will predict the generation of new restriction sites and the expected outcome of diagnostic digests.

This is an extraordinarily powerful feature. It should be noted that the program is very picky regarding fragment specifications and the order in which information is entered: if you don't do everything in the proper order (or if the fragments are in the wrong orientation), it generates cryptic error messages. Some of the construction schemes appear less than optimal, but they all appear to be workable and all serve as excellent starting points. Considering that this feature alone could save weeks of unproductive time in the laboratory, it is well worth the time needed to learn to use the program correctly.

In keeping with its purpose as a vector-construction program, Vector NTI includes a comprehensive module for designing PCR primers, sequencing primers, and hybridization probes. It also provides a nice electrophoresis simulation feature. You can rapidly simulate the electrophoresis of a restriction digest using agarose or acrylamide medium with several standard buffers and sets of running conditions. This is particularly useful in defining conditions under which to separate two closely migrating fragments. These features nicely enhance the molecular-design aspects of Vector NTI, providing a fairly comprehensive set of capabilities for vector construction and analysis.

Vector NTI also generates very nice graphics and tables. The quality of the printed graphics is very high. Some of the graphics can be edited, but this requires navigation through multiple dialogs and menus, making it difficult to achieve a desired effect. The graphics can be exported through the clipboard or as a file (in WMF or PICT formats).

Limitations

The Macintosh version of Vector NTI appears to have been a hasty port of the Windows version. There are numerous cosmetic problems, such as disappearing menus, words running together, mangled Save dialogs and erratic screen updates. Fortunately, the program rarely crashed. The Windows version of Vector NTI showed none of the cosmetic problems displayed by the Macintosh version.

The interfaces of the Macintosh and Windows versions are virtually identical. Unfortunately, many of the interface elements follow neither Macintosh nor Windows conventions. This makes it much more difficult to remember how to use the program after a brief absence. Vector NTI is very complex, making this is a significant problem. For example, many windows lacked a File menu. The copy and cut commands required different keystrokes in different windows. Opening and saving documents was accomplished using a variety of different commands in at least three different menus. Undo was only available for a few drawing functions. In BioPlot and AlignX, closing a window would quit the program. Many commands are executed using buttons in toolbars, but the button graphics are poorly designed and rarely informative. For example, an icon with either an arrow or magnifying glass is used to open different types of documents. There are no visual indications to suggest which panes in the display windows are active, and changing panes alters some of the menu items. Because of the poor program organization, the user usually must hunt through the menus and buttons to find a given command. Fortunately, the window interface itself is well designed, and most important commands are available by control-clicking to bring up a context-dependent menu.

The functions of the AlignX and BioPlot modules are very simple and can all be performed in very similar fashion through the usual Internet sites. For example, I tested the ability of AlignX to find homologies among a series of related hormones and it performed well, yielding the same result as the BCM server. The Macintosh version of AlignX displays serious performance and cosmetic problems. Alignment of the proteins in the included demo file (Protein.apr) required less than 2 seconds for the Windows version of AlignX, but over 2 minutes for the Macintosh version. Scrolling in the Alignment Editor garbles the sequences in the Macintosh version.

Although annotated files in the Vector NTI database are a snap to work with, entering these files can sometimes be difficult. For example, raw vector-sequence files, such as those obtained from many vendors, contain no motif information. With Vector NTI this information must be entered manually, because the program has no facility for identifying structural or functional motifs. You can easily identify all files in your database that contain Amp^r, for example, but there is no simple mechanism for locating Amp^r within a raw sequence file. Similarly, the AlignX package performs ClustalX analysis admirably, but lacks a simple alignment function. The program cannot find proteolytic sites or potential glycosylation sites in proteins. It is difficult to perform simple restriction digests, and you cannot specify codon usage for translation. Creating custom features or motifs is cumbersome, requiring the user to navigate multiple dialog boxes and scrolling lists. Preparing an annotated vector file typically takes several hours of work and is only possible if you know beforehand the sequences that define a given feature. Annotating the files is essential because many of the features of Vector NTI (such as molecular design aspects) requires the prior identification of structural features of the molecules.

Access to the Internet through Vector NTI is clunky. Basically, you chose the files you want to analyze online and Vector NTI launches your Web browser and pastes your sequence data into the data input window automatically. The program also allows you to transfer search results easily back into Vector NTI, but you must remember to save the file into the database manually. Closing the file window will erase your data without any warning from the program. Many of the Internet links in the evaluated copy of Vector NTI were out of date. Although the company claimed that a current set of links (tools) was available on its Web site, they were not available as of this writing.

File sharing among users is limited. InforMax offers an optional shared database that can be opened in a separate Database Explorer window by multiple users over a network. It allows users to log in and simply drag and drop sequences to and from the file. Each copy of Vector still retains its own local database. The shared database can be accessed by both Mac and PC users.

Although the suite is easy to install, the installation program places all the files in a single folder. Except for Vector NTI, the names of the installed programs are different than described in the documentation, making it difficult to locate them.

Vector NTI Suite has an onerous copy-protection scheme. The program is permanently locked to a single machine. Thus, if you have five people in the lab who want to use the program, you must either buy five copies of the program or install it on a common workstation. Other programs in this class usually use a "dongle" device that allows a single copy of the program to be transferred to a different machine. If you want to upgrade your computer hardware, you must contact InforMax and work out an arrangement to erase the program from the original machine and transfer it to the new machine.

For the construction and modification of custom vectors and genes, Vector NTI is far more comprehensive than any other program in its class. The database aspects of Vector NTI are also superior to those found in similar programs. However, it is relatively difficult to perform simple analyses with Vector NTI. To find restriction sites or to compare two sequences, simple packages such as GeneJockey or DNA Strider (for the Macintosh) or GeneWorks (for Windows) are better choices and are much less expensive. MacVector and Omiga for Windows are easier to use and are much more comprehensive in their support of protein analysis and identification of structural motifs, although they lack the vector design features of Vector NTI. MacVector is better integrated with the Internet, allowing BLAST searches from directly within the program. Most of the analysis features of Vector NTI, other than the design and construction aspects, are readily obtained over the Internet. It is probably most appropriate to consider Vector NTI to be a specialized sequence analysis package. It appears to be aimed primarily at the "power molecular biologist" who is designing a large number of expression vectors, rather than the average biologist who wants to express an occasional gene.

Technical Support and Documentation

Technical support by phone is fast, friendly, and very useful. This is particularly important for Vector NTI, because the printed documentation is so poor as to be almost useless. The tutorials were somewhat helpful, but they contained frequent errors, such as misidentifying the location of menu items. Many operations described in the tutorial simply do not work. The online help is not much better. Many dialog boxes have a help buttons, but the usual effect of clicking them is an alert box that states, "Help not available on this topic." The rest of the documentation consists of prosaic descriptions of what the program features accomplish. For example, Vector NTI provides a set of tools for designing PCR primers. The dialog boxes provide for dozens of different options, such as DG Temperature, palindromes, and primer-primer complementarity, but there is nothing in the documentation to explain the purpose of these options. The entire explanation of the topic "Primer-primer complementarity" is the statement "Check Permitted box if primer-primer complementarity is permitted, and enter the minimum permitted value for duplex free energy." No references are provided and no explanation of utility is provided for any feature of the program. This is particularly disappointing considering the program's complexity and high cost.

Utility

Best for "power molecular biologists" who spend a considerable amount of time designing and building expression vectors and modified genes.