What types of genome maps are there?

The two main kinds of genome maps are known as genetic-linkage maps and physical maps. Just like the various types of earth maps, they contain different types of information and have different strengths and weaknesses. For example, on a political map of the United States, Colorado appears as a featureless rectangle that gives no hint to the state's dramatic scenery. On a physical map, Colorado is a stunning collection of mountain peaks, high plateaus, and open prairie—natural features that don't tell you anything about where the state's boundaries are.

Image of crossing over occuring in chromosomes

Genetic-linkage mapping. Genetic-linkage maps illustrate the order of genes on a chromosome and the relative distances between those genes. Originally, these maps were made by tracing the inheritance of multiple traits, such as hair color and eye color, through several generations.

Genetic-linkage mapping is possible because of a normal biological process called crossing over, which occurs during meiosis—a type of cell division for making sperm and egg cells. During one stage of meiosis, chromosomes line up in pairs along the center of a cell, where they sometimes "stick" to each other and exchange equivalent pieces of themselves. This sticking and exchanging is called crossing over, and is a relatively common event: on average, a chromosome pair undergoes crossing over about 1.5 times during the formation of each sex cell in humans.

For example, imagine a man who has one chromosome with brown-eye and brown-hair genes, and another chromosome with blue-eye and blonde-hair genes in his cells. Usually, his sperm cells will have either brown-eye and brown-hair genes, or blue-eye and blonde-hair genes. But if crossing over occurs, the man will produce one sperm cell with brown-eye and blonde-hair genes, and another with blue-eye and brown-hair genes.

In short, crossing over produces chromosomes with new combinations of genes—and offspring, called recombinants, with new combinations of traits not seen in either parent. Generally the closer two genes are on a chromosome, the less likely they are to be separated by crossing over.

This means that traits that are inherited together most often are probably influenced by genes that are close to each other on a chromosome. On the other hand, traits that are inherited together less often are probably influenced by genes that are farther apart. Thus, by following several traits through generations and recording how often recombinants occur, one can map the relative position of corresponding genes.

A. H. Sturtevant, then a student at Columbia University, made the first genetic-linkage map of fruit fly genes in 1913—decades before scientists even knew that genes are made of DNA. He found, for example, that leg length was inherited with eye color more often than with wing length, and that wing length was inherited with eye color more often than with leg length. Thus, he concluded, the gene for eye color must be between the genes for wing length and leg length in the fruit fly genome.

But because the frequency of crossing over varies at different places in the genome, these early genetic-linkage maps only gave the relative positions of genes, not their physical locations in the genome or their actual distances, in DNA base pairs, from each other. In addition, the maps were based on the inheritance of traits, not genes, and this limited the possible landmarks to characteristics that were visible or measurable in some way.

Today, scientists make genetic-linkage maps by tracing the inheritance of certain DNA sequences the same way they once traced the inheritance of visible traits. The genome contains many places where the DNA sequence varies from person to person (where I have AAC you have AAG, for example). These sequence variations, or polymorphisms, make up many of the landmarks on modern genetic-linkage maps and enable scientists to anchor genes to their true physical locations in the genome.

Physical mapping. Physical maps, by contrast, always give the physical, DNA-base-pair distances from one landmark to another.

In the late 1970s, scientists developed new and efficient ways of cutting the genome up into smaller pieces in order to study it. Around the same time they made the first physical maps by using the overlapping DNA sequences at the ends of the genome pieces to help them keep track of where the pieces came from. (The process had a lot in common with the assembly step of genome sequencing.) In other words, a physical map was simply an ordered set of DNA pieces.

This worked okay for a while, but then more and more scientists started getting interested in the genome, and they were all cutting it up in different ways and building physical maps from different sets of DNA pieces. The scientists couldn't share information with each other, because they each had maps written, in effect, in a different language. Moreover, the landmarks they were using weren't necessarily unique—that is, a landmark could appear in more than one place in the genome, so finding a landmark didn't necessarily tell you where you were.

Today, genome scientists use landmarks known as STSs to help them find their way around the genome. Each STS, or "sequence-tagged site," is a unique DNA sequence—one that is found in only one place in the genome—and is a few hundred base pairs long. Some STSs are parts of genes, but an STS can come from anywhere in the genome as long as it is unique.

No matter how you cut up a genome, STSs will tell you where the pieces belong. For example, once an STS has been added to the genome map, you can figure out whether any piece of DNA contains that STS landmark. If it does, you know exactly where in the genome that piece of DNA belongs.

Scientists began to use STSs to construct maps in the late 1980s. Thus, physical maps have evolved from an ordered set of DNA pieces (which can be used only by people who have the same set of pieces) into a set of landmarks based on unique DNA sequences (which can be used by any scientist the world over).

What are the maps like that are being made now? What will the genome map be like when it is done?

The genome's cartographers are now making maps that combine features of both genetic-linkage and physical maps. For example, new methods of genetic-linkage mapping based on polymorphisms enable scientists to place landmarks at the proper physical locations on the chromosomes. Meanwhile, some STSs on physical maps are parts of genes and thus give the sort of information that appears on genetic-linkage maps.

As mapping techniques advance, scientists try to create maps with more landmarks that are more closely, evenly, and accurately spaced. But in contrast to DNA sequencing, which has become increasingly automated, genome mapping still can only be accomplished by experienced scientists. And even the most expert mapper may run into difficulties finding the desired number and type of landmarks with the desired spacing. This means that although maps keep improving, the work is slow and there is still a long way to go.

It's almost as difficult (and arbitrary) to define a "complete" genome map as it is to define a "complete" genome sequence. One definition of "complete" is a map that includes the sequence and location of all of an organism's genes. Such maps currently exist for more than 150 organisms, most of them viruses with small genomes.

.  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  . 

Updated on January 15, 2003

Genome News Network is an editorially independent online publication of the J. Craig Venter Institute. © 2000 - 2004 J. Craig Venter Institute. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.