Recently, the new term in silico has become a common reference to biological studies carried out in the computer, joining the traditional terms in vivo and in vitro to describe the location of experimental studies.
For nonbiologists, in vitro means "in glass," that is, in the test tube; in vivo means "in life," that is, in a living organism. The term in silico stems from the fact that most computer chips are made primarily of silicon. Personally, I prefer a term such as in algorithmo, since there are plenty of ways to compute that don't involve silicon, such as the intriguing processes of DNA computing, quantum computing, optical computing, and more.
The large amount of biological data available online has brought biological research to a situation somewhat similar to physics and astronomy. Those sciences have found that experiments in modern equipment produce huge amounts of data, and the computer isn't only invaluable but necessary for exploring the data. Indeed, it's become possible to simulate experiments entirely in the computer. For instance, an early use of computer simulation in physics was in modeling the acoustics of a concert hall and then experimenting with the results by changing the design of the hall—clearly a much cheaper way to experiment than by building dozens of concert halls!
A similar trend has been occurring in biology since computers were first invented, but this trend has sharply accelerated in recent years with the Human Genome Project and the sequencing of the DNA of many organisms. The experimental data that has to be collected, searched, and analyzed is often far too large for the unaided biologist, who is now forced to rely on computers to manage the information.
Beyond the storage and retrieval of biological data, it's now possible to study living systems through computer simulation. There are standard and accepted studies done routinely on computers that access the genes of humans and of several other organisms. When the sequence of some DNA is determined, it can be stored in the computer, and programs can be written to identify restriction sites, perform restriction digests and create restriction maps (see Chapter 9). Similarly, gene-finding programs can take sequenced DNA and identify putative exons and introns. (Not perfectly, as of this writing, and results differ for different organisms.) Models of cellular processes exist in which it is possible to study for example, the effect of a change in the regulation of a gene.
Today, microarray technology (incorporating glass slides spotted with thousands of samples that can be probed, usually with the aid of robotics) can assess the levels of expression of thousands of genes with one laboratory run. Computers are helping to unravel the complex interactions between genes. We hope to find, for example, all sets of genes related by virtue of their protein products as part of a biochemical pathway in the cell. Microarrays generate a large volume of data. This data needs to be stored, compared with other experimental data, and analyzed on the computer.
On my first day as a programmer at Bell Labs Research, my boss told me that his simulations could now be computed so fast—overnight—that it was creating a problem for him. There wasn't enough time to think about the last simulation! Nevertheless, and despite all the attendant headaches and pitfalls of computers, their use to simulate experiments is proving to be beneficial in biology.
1.4 Limits to Computation
Some of the most interesting results of computer science demonstrate certain limits to human knowledge. There are many open problems in biology, and one hopes that applying more computer power to them may help solve them. But this isn't always possible, because some problems can be shown to be unsolvable; that is, they can't be solved by any program. Furthermore, some problems may be solvable, but as the size of the problem grows, they get practically impossible to solve. These problems are called intractable , or NP-complete. Even a million computers, each a million times more powerful than the most powerful computer existing today, could take perhaps a billion years to compute the answer to such an intractable problem.
Now the chances are that you're not going to get stung by an unsolvable or intractable problem. It can happen, but it's relatively rare. I mention them more as a point of interest than as a practical concern to the beginning programmer. But as you attempt more complex programs down the road, these limitations, and especially the intractable nature of several biological problems, can have a practical impact on your programming efforts.
Chapter 2. Getting Started with Perl
Perl is a popular programming language that's extensively used in areas such as bioinformatics and web programming. Perl has become popular with biologists because it's so well-suited to several bioinformatics tasks.
Perl is also an application, just like any other application you might install on your computer. It is available (at no cost) and runs on all the operating systems found in the average biology lab (Unix and Linux, Macintosh, Windows, VMS, and more). The Perl application on your computer takes a Perl language program (such as one of the programs you will write in this book), translates it into instructions the computer can understand, and runs (or "executes") it.
 An operating system manages the running of programs and other basic services that a computer provides, such as how files are stored.
So, the word Perl refers both to the language in which you will write programs and to the application on your computer that runs those programs. You can always tell from context which meaning is being used.
Every computer language such as Perl needs to have a translator application (called an interpreter or compiler) that can turn programs into instructions the computer can actually run. So the Perl application is often referred to as the Perl interpreter, and it includes a Perl compiler as well. You will often see Perl programs referred to as Perl scripts or Perl code. The terms program, application, script, and executable are somewhat interchangeable. I refer to them as "programs" in this book.
2.1 A Low and Long Learning Curve
A nice thing about Perl is that you can learn to write programs fairly quickly; in essence, Perl has a low learning curve. This means you can get started easily, without having to master a large body of information before writing useful programs.
Perl provides different styles of writing programs. Since these are beyond the scope of this book, I won't go into details, except to mention the popular style called imperative programming that you'll learn in this book. The equally popular style called object-oriented programming is also well-supported in Perl. Other styles of programming include functional programming and logic programming.
Although you can get started quickly, learning all of Perl will certainly take awhile, if that's your goal. Most people learn the basics, as presented in this book, and then learn additional topics as needed.
Let's get a few elementary definitions out of the way:
What is a computer program?
It's a set of instructions written in a particular programming language that can be read by the computer. A program can be as simple as the following Perl language program to print some DNA sequence data onto the computer screen:
The Perl language programs are written and saved in files, which are ways of saving any kind of data (not only programs) on a computer. Files are organized hierarchically in groups called folders on Macintosh or Windows systems or directories in Unix or Linux systems. The terms folder and directory will be used interchangeably.
What is a programming language?
It's a carefully defined set of rules for how to write computer programs. By learning the rules of the language, you can write programs that will run on your computer. Programming languages are similar to our own natural, or spoken languages, such as English, but are more strictly defined and specific to certain computer systems. With a little bit of training, it's not difficult to read or write computer programs. In this book you'll write in Perl; there are many other programming languages.
A program that a programmer writes is also called source code, or just source or code. The source code has to be turned into machine language, a special language the computer can run. It's hard to write or read a machine language program because it's all binary numbers; it's often called a binaryexecutable. You use the Perl interpreter (or compiler) to turn a Perl program into a running program, as you'll see later in this chapter.
What is a computer?
Okay, silly question. It's that machine you buy in computer stores. But actually, it's important to have a clear idea of what kind of machine a computer is. Essentially, a computer is a machine that can run many different programs. This is the fundamental flexibility and adaptability that makes the computer such a useful and general-purpose tool. It's programmable; you will learn how to program it using the Perl programming language.
2.2 Perl's Benefits
The following sections illustrate some of Perl's strong points.
2.2.1 Ease of Programming
Computer languages differ in which things they make easy. By "easy" I mean easy for a programmer to program. Perl has certain features that simplifies several common bioinformatics tasks. It can deal with information in ASCII text files or flat files, which are exactly the kinds of files in which much important biological data appears, in the GenBank and PDB databases, among others. (See the discussion of ASCII in Chapter 4; Genbank and PDB are the subjects in Chapter 10 and Chapter 11.) Perl makes it easy to process and manipulate long sequences such as DNA and proteins. Perl makes it convenient to write a program that controls one or more other programs. As a final example, Perl is used to put biology research labs, and their results, on their own dynamic web sites. Perl does all this and more.
Although Perl is a language that's remarkably suited to bioinformatics, it isn't the only choice nor is it always the best choice. Other programming languages such as C and Java are also used in bioinformatics. The choice of language depends on the problem to be programmed, the skills of the programmers, and the available system.
2.2.2 Rapid Prototyping
Another important benefit of using Perl for biological research is the speed with which a programmer can write a typical Perl program (referred to as rapid prototyping). Many problems can be solved in far fewer lines of Perl code than in C or Java. This has been important to its success in research. In a research environment there are frequent needs for programs that do something new, that are needed only once or occasionally, or that need to be frequently modified. In Perl, you can often toss such a program off in a few minutes or a few hours work, and the research can proceed. This rapid prototyping ability is often a key consideration when choosing Perl for a job. It is common to find programmers familiar with both Perl and C who claim that Perl is five to ten times faster to program in than C. The difference can be critical in the typical understaffed research lab.
2.2.3 Portability, Speed, and Program Maintenance
Portability means how many types of computer systems the language can run on. Perl has no problems there, as it's available for virtually all modern computers found in biology labs. If you write a DNA analyzer in Perl on your Mac, then move it to a Windows computer, you'll find it usually runs as is or with only minor retrofitting.
Speed means the speed with which the program runs. Here Perl is pretty good but not the best. For speed of execution, the usual language of choice is C. A program written in C typically runs two or more times faster than the comparable Perl program. (There are ways of speeding up Perl with compilers and such, but still... .)
In many organizations, programs are first written in Perl, and then only the programs that absolutely need to have maximum speed are rewritten in C. The fact is, maximum speed is only occasionally an important consideration.
Programming is relatively expensive to do: it takes time, and skilled personnel. It's labor-intensive. On the other hand, computers and computer time (often called CPU time after the central processing unit) are relatively inexpensive. Most desktop computers sit idle for a large part of the day, anyway. So it's usually best to let the computer do the work, and save the programmer's time. Unless your program absolutely must run in say, four seconds instead of ten seconds, you're okay with Perl.
Program maintenance is the general activity of keeping everything working: such activities as adding features to a program, extending it to handle more types of input, porting it to run on other computer systems, fixing bugs, and so forth. Programs take a certain amount of time, effort and cost to write, but successful programs end up costing more to maintain than they did to write in the first place. It's important to write in a language, and in a style, that makes maintenance relatively easy, and Perl allows you to do so. (You can write obscure, hard-to-maintain code in Perl, as in other languages, but I'll give you pointers on how to make your code easy for other programmers to read.)
Perl, like almost all popular software, has gone through much growth and change over the course of its nearly 15-year life. The authors—Larry Wall and a large group of cohorts—publish new versions periodically. These new versions have been carefully designed to support most programs written under old versions, but occasionally some major new features are added that don't work with older versions of Perl.
This book assumes you have Perl Version 5 or higher installed. If you have Perl installed on your computer, it's likely Perl 5, but it's best to check. On a Unix or Linux system, or from an MS-DOS or MacOS X command window, the perl -v command displays the version number, in my case, Version 5.6.1. The number 5.6.1 is "bigger" than 5; that means it's okay. If you get a smaller number (very likely 4.036), you have to install a recent version of Perl to enable the majority of programs in this book to run as shown.
What about future versions? Perl is always evolving, and Perl Version 6 is on the horizon. Will the code in this book still work in Perl 6? The answer is yes. Although Perl 6 is going to add some new things to the language, it should have no trouble with the Perl 5 code in this book.
2.3 Installing Perl on Your Computer
The following sections provide pointers for installing Perl on the most common types of computer systems.
2.3.1 Perl May Already Be Installed!
Many computers—especially Unix and Linux computers—come with Perl already installed. (Note that Unix and Linux are essentially the same kind of operating system; Linux is a clone, or functional copy, of a Unix system.) So first check to see if Perl is already there. On Unix and Linux, type the following at a command prompt:
$ perl -v
If Perl is already installed, you'll see a message like the one I get on my Linux machine:
This is perl, v5.6.1 built for i686-linux
Copyright 1987-2001, Larry Wall
Perl may be copied only under the terms of either the Artistic License or the
GNU General Public License, which may be found in the Perl 5 source kit.
Complete documentation for Perl, including FAQ lists, should be found on
this system using 'man perl' or 'perldoc perl'. If you have access to the
Internet, point your browser at http://www.perl.com/, the Perl Home Page.
If Perl isn't installed, you'll get a message like this:
perl: command not found
If you get this message, and you're on a shared Unix system at a university or business, be sure to check with the system administrator, because Perl may indeed be installed, but your environment may not be set to find it. (Or, the system administrator may say, "You need Perl? Okay, I'll install it for you.")
On Windows or Macintosh, look at the program menus, or use the find program to search for perl. You can also try typing perl -v, at an MS-DOS command window or at a shell window on the MacOS X. (Note that the MacOS X is a Unix system!)
2.3.2 No Internet Access?
If you don't have Internet access, you can take your computer to a friend who has access and connect long enough to install Perl. You can also use a Zip drive or burn a CD from a friend's computer to bring the Perl software to your computer. There are commercial shrink-wrapped CDs of Perl available from several sources (ask at your local software store) and several books such as O'Reilly's Perl Resource Kits, include CDs with Perl.
Apart from installing Perl, you don't need Internet access for everything in this book. If you want to do the exercises while commuting on the train, or whatever, it can certainly be done. Apart from installing Perl, the main use of the Internet for this book is to download its examples from the book's web site without having to type them; to download and try the exercises; to explore biological data from various biological databases; and to access Perl documentation, if it's not installed on your machine.
Know that if you want to do bioinformatics, the Internet is a practical necessity. You can learn programming fundamentals from this book without an Internet connection, but you will need Internet access to download bioinformatics software and data.
Perl is an application, so downloading and installing it on your computer is pretty much the same as installing any other application.
The web site that serves as a central jumping off point for all things Perl is http://www.perl.com/. The main page has a Downloads clickable button that guides you to everything you need to install Perl on your computer. At the Downloads page, there's a Getting Help link and other links. So even if the information in this book becomes outdated, you can visit the Perl site and find all you need to install Perl.
Downloading and installing Perl is usually quite easy, in fact, the majority of the time it's perfectly painless. However, sometimes you may have to put some effort into getting it to work. If you're new at programming, and you run into difficulties, you should ask for help from a professional computer programmer, administrator, teacher, or someone in your lab who already programs in Perl.
So, in a nutshell, here are the basic steps for installing Perl on your computer:
Check to see if Perl is already installed; if so, check the that version is at least Perl 5.
Get Internet access and go to the Perl home page at http://www.perl.com/.
Go to the Downloads page and determine which distribution of Perl to download.
When downloading from the http://www.perl.com site, you need to choose between binary or source-code distributions of Perl. The best choice for installing Perl on your computer is to get an already made binary version of the program, because it's the easiest to install. However, if no binary is available, or if you want to control the various options of your Perl installation, you can get the source code for Perl, which is itself written in the C programming language. You then compile it using a C compiler. But try to find a binary for your particular computer's operating system; compiling from source code can be complicated for beginners.
The next sections provide specific installation instructions for specific platforms.
188.8.131.52 Unix and Linux
If Perl isn't installed on your Unix or Linux machine, first try to find a binary to install. At the Downloads page of http://www.perl.com, you'll see the subheading Binary Distributions. Select Unix or Linux, and then see if your particular flavor of operating system has a binary available. Several versions are available, and the web-site instructions should be enough to get Perl installed once you've downloaded the binary. Most versions of Linux maintain up-to-date Perl binaries on their web sites. For instance, if you have a Red Hat Linux system, you need to identify which version of the system you have (by typing uname -a) and then get the appropriate rpm file to download and install. Red Hat has an rpm for Perl that Red Hat Linux users can install by typing:
rpm -Uvh perl.rpm
(the actual name of the perl.rpm file varies).
If no binary version of Perl is available for your flavor of Unix or Linux, you must compile Perl from its source code. In this case, starting from the Perl web page, click on the Downloads button and then select Source Code Distribution. The source code has an INSTALL file with instructions that guide you through the process of downloading the source code, installing it on your system, compiling the source code into a binary, and finally installing the binary.
As mentioned previously, compiling from source code is a considerably longer process than installing an already made binary, and requires a bit more reading of instructions, but it usually works quite well. You will need a C compiler on your computer to install from source code. Nowadays, some Unix systems ship without a complete C compiler. Linux will always have the free C compiler called gcc installed, and you can also install gcc on any Unix (or Windows, or Mac) system that lacks a C compiler.
The MacPerl installation steps are clearly explained on the MacPerl web page, http://www.macperl.com/ (which you can also get to from the Perl web page and its Downloads button). Here's a very brief overview.
From the MacPerl page, click on Get MacPerl, and follow the directions to download the application. It will appear on your desktop. Double-click it to unstuff it. If you don't have Aladdin Stuffit Expander (most Macs already do), this won't work, and you'll have to go to http://www.aladdinsys.com to download and install Stuffit.
MacPerl can be installed as a standalone application under the MacOS Finder or as a tool under the Macintosh Programmer's Workbench; you will probably want the standalone application. Perl Version 5 is available for MacOS 7.0 and later. Details about which Perl version is available for your particular hardware and MacOS version are available at the MacPerl web page.
Several binaries for different Windows versions are available. Since Windows is closely coupled with Intel 32-bit chips, these binaries are often called Wintel or Win32 binaries. The current standard Perl distribution is ActivePerl from ActiveState, at http://www.activestate.com/ActivePerl/, where you can find complete installation directions. You can also get to ActivePerl via the Downloads button from the Perl web site. Under the subheading Binary Distributions, go to Perl for Win32, and then click on the ActivePerl site.
From the ActiveState web site's ActivePerl page, click the Downloads button. You can then download the Windows-Intel binary. Note that installing it requires a program called Windows Installer, which is available at ActivePerl if it's not already on your computer.