Mark Patterson

Multifactorial genetics: Wake-up call for genome scanners

Meiosis-specific processes, such as pairing of homologous chromosomes and recombination, are highly conserved, nevertheless there have been conflicting theories about the sequence of events that leads up to recombination. In yeast, it is well established that meiotic exchange can take place in the absence of the synaptonemal complex (SC)  a proteinaceous ‘glue’ that holds sister chromatids together. However, in a recent issue of Genes and Development, Page and Hawley provide new evidence that, in flies, the SC is essential for the initiation of recombination, and that there might be some unexpected and fundamental differences between meiosis in flies and in yeast. Previous evidence from Drosophila mutants has shown that the relationship between recombination and the SC is not the same as it is in yeast. For example, the mei-W68 mutant has a normal SC but does not undergo recombination and the c(3)G mutant eliminates both the SC and recombination. To investigate whether recombination depends on the SC, Page and Hawley have cloned c(3)G and shown that it encodes an essential component of the SC. As expected, C(3)G localizes between the paired homologues, and this localization is altered in mutants that disrupt the SC, confirming its role as an integral part of the SC. Using C(3)G as a marker, Page and Hawley looked at the SC in mutants in which meiotic exchange is defective. One class of these mutants has reduced frequency of exchange, and the distribution of crossover sites is biased against distal parts of the chromosomes  a phenomenon known as polar effect. Because C(3)G is mislocalized in these mutants, the authors conclude that incorrect SC assembly might, at least partially, account for the polar effect. But even in normal cells, crossovers are not randomly distributed along the chromosome because of crossover interference  the suppression of crossovers in the immediate neighbourhood of an established crossover point. In yeast, incorrect assembly of the SC and incomplete pairing of the homologues abolishes interference, but in Drosophila, Page and Hawley show that when the SC is absent because C(3)G is mislocalized or non-functional, interference is essentially unaffected. The picture that emerges is that, in contrast to yeast, which might use the initiation of recombination to align homologous chromosomes, and the SC just to stabilize their pairing, Drosophila needs the SC for the initial alignment of the homologues, without which recombination will not be initiated. It also seems that these two organisms have different ways of controlling interference  in yeast this process is SC dependent, whereas in flies it is SC independent. It remains to be seen which way is more common among other organisms. Magdalena Skipper

Linkage disequilibrium. Highs and lows.

RICHARD YOUNG MASSACHUSETTS INSTITUTE OF TECHNOLOGY, USA Representing the human genome as a sequence of bases or genes does not capture its higher level organization. But stepping back from the sequence can reveal a complex terrain with undulating patterns of sequence and structure — patterns that reflect differences such as base composition and chromosome banding. For one region in the genome, Eisenbarth et al. have found an association between base composition and linkage disequilibrium (LD) — an observation that has important implications for mapping the genes that underlie disease. When two loci are close together, recombination between them is rare. Alleles at the two loci tend to occur together in a population more frequently than would be expected if they segregated randomly. The loci are said to be in LD and this provides a means to scan the genome for disease-susceptibility genes — if sufficient markers are used, it should be possible to detect LD between a marker and a candidate disease gene, and then home in on the gene itself. But recombination rates (and therefore LD) vary across the genome, so how do you know how many markers to use? The neurofibromatosis gene, NF1, is in a region of the human genome with variable LD and lies at the junction of two isochores — stretches of the genome associated with different levels of G+C content. Eisenbarth et al. show that the isochore junction coincides with a sharp transition in LD — high G+C is associated with low LD. From the degree of LD, they estimate that recombination rates on either side of the junction vary by around 100-fold, and each of the two regions extends for several hundred kilobases. If this observation is repeated for other parts of the genome, G+C content could help to predict levels of LD. This will not only inform strategies to screen for LD in the genome, but will also provide an indication of the size of the region that needs to be searched once LD is identified. So, before you embark on an expedition to find genes associated with a particular disease, it could pay to know your genomic terrain. Mark Patterson

Mining gene expression data.

NATURE REVIEWS | GENETICS VOLUME 1 | DECEMBER 2000 | 165 Homophila Despite Homophila’s spooky homepage, human geneticists curious to know what their disease gene does in Drosophila have nothing to fear. Ethan Bier and his colleagues at UCSD have compared the gene sequences entered in the Online Mendelian Inheritance in Man (OMIM) database to the genes, EST or genomic sequences in Flybase, the Drosophila sequence database. There is a story behind the creation of this web site. After finding that 74.5% of 909 distinct human disease genes have close homologues in fruitflies, the researchers scribbled the gene names and their corresponding syndromes on cards and handed them to a pathologist, who dutifully placed them into piles according to the nature of the disorder. This evolved into Homophila, a site where human disease genes and their fly homologues can be searched according to keyword, human disease name, gene name or OMIM entry number. For instance, typing in ‘hypertension’ leads to a results table showing, on the left, a description of the human disorder, the human gene symbol (for example, for the angiotensin II receptor) and related online references; on the right, the Drosophila protein sequence matches. In case you’re daunted by the prospect of trundling through fruitfly data, a link from the fly gene of interest to GadFly (Genome Annotation Database of Drosophila) leads you to all that is known about the gene. There’s more to come, as the site curators promise a facility to match disease phenotypes that are common to humans and flies. As molecular signalling pathways are more completely described in flies compared with humans, human disease genes could be cloned on the basis of fly mutant phenotypes that are typical of a particular pathway. The site is updated monthly, so keep a lookout. Tanita Casci The use of microarrays to monitor the transcription of thousands of genes under multiple conditions or in multiple cell lines is generating a massive and growing amount of valuable data. But there is a pressing need for more and better analysis tools. Two recent papers report new approaches and show how different methods of data mining can yield new information. Both papers use gene expression data related to cancer biology. One way of analysing microarray data is to look for groups of genes whose expression patterns are similar across many experiments. The co-regulated genes within such clusters are often found to have related functions. Getz et al. started with the idea that some gene clusters might be masked by transcriptional ‘noise’ from genes outside the cluster, or if the genes are co-regulated in only a subset of the experiments. So the authors developed an algorithm called coupled two-way clustering that breaks down the total dataset into subsets of genes and samples that can reveal significant clusters. Two previously published datasets were used by Getz et al. The first comprised 72 samples of two types of acute leukaemia — acute lymphoblastic leukaemia (ALL) and acute myeloid leukaemia (AML). After applying their analysis, they identified 84 clusters. One of the clusters (comprising 60 genes) separated the samples into AML and ALL. Another cluster (of 28 genes) split the AML patients into those who had received treatment and those who had not. The second dataset used by Getz et al. comprised 40 colon cancer samples and 22 controls. Their analysis was able to split the group into the normal and diseased samples using one of the clusters of genes, and another cluster partitioned the samples according to a difference in the methodology used for RNA preparation. Overall, the method does generate meaningful clusters that are not detected when the whole dataset is analysed. The task is now to examine the unexplained clusters to look for biological significance. Butte et al. used an entirely different approach to mine data from two different datasets. The data concerned 60 cell lines established by the National Cancer Institute and used since 1989 to screen anticancer agents. The first dataset comprised the transcript levels for several thousand genes in each cell line. The second dataset comprised the GI50 (the level of anticancer agent required to achieve 50% growth inhibition) for several thousand agents on each cell line. The aim was to look for significant correlation between every possible pair of agents and genes. Correlations were summarized diagrammatically in networks and 202 such networks were found. Many expected associations were found between structurally related anticancer agents, and networks were also identified that linked genes of related function. Only one association was found between an agent and a gene — the GI50 for a thiazolidine carboxylic acid derivative increased with the expression of the gene LCP1, which encodes an actin-binding protein. Once again, the networks need to be analysed further to uncover the biological meaning. Among the advantages of this method are that individual genes or agents can be linked more than once, and that negative correlations can be found just as easily as positive correlations. These two papers expand the range of tools for analysis of transcript profile data, and expose further seams for would be data-miners. Mark Patterson References and links ORIGINAL RESEARCH PAPERS Getz, G. et al. Coupled two-way clustering analysis of gene microarray data. Proc. Natl Acad. Sci. USA 97, 12079–12084 (2000) | Butte, A. J. et al. Discovering functional relationships between RNA expression and chemotherapeutic susceptibility using relevance networks. Proc. Natl Acad. Sci. USA 97, 12182–12186 (2000) WEB SITES Computational physics group, Weizmann Institute | Molecular pattern recognition, Whitehead Institute Mining gene expression data B I O I N F O R M AT I C S WEB WATCH

Genome evolution: A primate example of positive selection

www.nature.com/reviews/genetics Segmental duplications are a common feature of the human genome and they can be very difficult to sequence. But they should not be ignored. Through careful analysis of one set of segmental duplications, Matthew Johnson and colleagues have discovered a new human gene family. Furthermore, this gene family seems to have been subjected to powerful positive selection. The authors focused on a 20-kb repeated segment that is confined to a 15-Mb region of chromosome 16. There are 15 copies of this segment, which have very high levels of sequence similarity. To study the evolutionary history of the duplicated segments, the authors identified the orthologous sequences in a series of primates. Their analysis showed that the segment is only present in one or two copies in Old World monkeys, such as the baboon. By contrast, in great apes such as gorillas, which are more closely related to humans, there are 9–30 copies of the segment. Overall, the segment seems to have been duplicated recently and independently in several primate lineages, after the divergence of humans and great apes from Old World monkeys. When the sequence of the human segment was used in a database search, the authors found that an expressed sequence lurks within the repeated segment, although no homologues could be found in other organisms. Surprisingly, sequence comparisons of the human repeats showed that the putative protein-coding regions are five times more divergent compared with the non-coding regions of the repeat. This indicates that the gene might be under positive selection for adaptive mutations. In support of this, the ratio of nonsynomous to synonymous amino-acid changes was significantly greater than 1, and on the basis of comparisons with the primate sequences, evidence of positive selection could be found during the divergence of the great ape and human lineages. Despite the relatively recent duplication events, the gene has undergone major evolutionary change. Positive selection for amino-acid substitutions indicates an important function, which, for this gene family, is as yet unknown. Nevertheless, these observations attest to the importance of duplication and divergence as key evolutionary mechanisms. So, although segmental duplications can be a positive nuisance for genome sequencers, they provide some fascinating clues about the history of our genome. Mark Patterson References and links ORIGINAL RESEARCH PAPER Johnson, M. E. et al. Positive selection of a gene family during the emergence of humans and African apes. Nature 413 , 514–519 (2001) WEB SITE Evan Eichler’s lab: http://genetics.gene.cwru.edu/eichler/ The positional cloning of a human disease gene is a landmark event. Good examples are provided by Bardet–Biedl syndrome (BBS), for which several genes have recently been identified (see Highlights, July). But these landmarks have now assumed greater significance, because some cases of BBS seem to require at least three mutant alleles. As a clear-cut example of ‘triallelic’ inheritance, BBS might therefore provide some general lessons about the interaction between genes, and its influence on phenotype. BBS is usually thought of as an autosomalrecessive disorder that is characterized by a broad and variable phenotype, involving retinal dystrophy, obesity, mental retardation, renal defects and polydactyly. BBS is also genetically heterogeneous — defects at several loci can cause the disease. So, when Katsanis et al. screened patients for mutations in two known BBS genes (BBS2 and BBS6), it was no surprise to find mutations in these genes in only some of the patients. What was more surprising was that some affected individuals carried three mutant alleles — for example, two in BBS2 and one in BBS6. The authors propose that BBS is inherited, at least in some individuals, in a triallelic fashion, although further BBS genes will need to be identified to determine the generality of this observation. In an accompanying Perspective, Arthur Burghes and colleagues, describe BBS inheritance as autosomal recessive with a modifier of penetrance, and liken BBS to other disorders in which modifier loci have been identified. However these unexpected findings are described, BBS provides a clear example of a phenotype that is determined by interactions between a small number of genes. BBS inheritance therefore lies somewhere between Mendelian and polygenic. As researchers move towards a more molecular understanding of BBS pathology, this disorder will become a valuable source of information about potential genetic interactions in more complex genetic disease. Mark Patterson References and links ORIGINAL RESEARCH PAPER Katsanis, N. et al. Triallelic inheritance in Bardet–Biedl syndrome, a Mendelian recessive disorder. Science 293, 2256–2259 (2001) FURTHER READING Burghes, A. H. M. et al. The land between Mendelian and multifactorial inheritance. Science 293, 2213–2214 (2001) | Nadeau, J. H. Modifier genes in mice and humans. Nature Rev. Genet. 2, 165–174 (2001) WEB SITES James Lupski’s lab: http://imgen.bcm.tmc.edu/molgen/lupski/ Bardet–Biedl syndrome links: http://www.kumc.edu/gec/support/laurmoon.html A bridge to complex disease H U M A N G E N E T I C S

A primate example of positive selection: Genome evolution

www.nature.com/reviews/genetics Segmental duplications are a common feature of the human genome and they can be very difficult to sequence. But they should not be ignored. Through careful analysis of one set of segmental duplications, Matthew Johnson and colleagues have discovered a new human gene family. Furthermore, this gene family seems to have been subjected to powerful positive selection. The authors focused on a 20-kb repeated segment that is confined to a 15-Mb region of chromosome 16. There are 15 copies of this segment, which have very high levels of sequence similarity. To study the evolutionary history of the duplicated segments, the authors identified the orthologous sequences in a series of primates. Their analysis showed that the segment is only present in one or two copies in Old World monkeys, such as the baboon. By contrast, in great apes such as gorillas, which are more closely related to humans, there are 9–30 copies of the segment. Overall, the segment seems to have been duplicated recently and independently in several primate lineages, after the divergence of humans and great apes from Old World monkeys. When the sequence of the human segment was used in a database search, the authors found that an expressed sequence lurks within the repeated segment, although no homologues could be found in other organisms. Surprisingly, sequence comparisons of the human repeats showed that the putative protein-coding regions are five times more divergent compared with the non-coding regions of the repeat. This indicates that the gene might be under positive selection for adaptive mutations. In support of this, the ratio of nonsynomous to synonymous amino-acid changes was significantly greater than 1, and on the basis of comparisons with the primate sequences, evidence of positive selection could be found during the divergence of the great ape and human lineages. Despite the relatively recent duplication events, the gene has undergone major evolutionary change. Positive selection for amino-acid substitutions indicates an important function, which, for this gene family, is as yet unknown. Nevertheless, these observations attest to the importance of duplication and divergence as key evolutionary mechanisms. So, although segmental duplications can be a positive nuisance for genome sequencers, they provide some fascinating clues about the history of our genome. Mark Patterson References and links ORIGINAL RESEARCH PAPER Johnson, M. E. et al. Positive selection of a gene family during the emergence of humans and African apes. Nature 413 , 514–519 (2001) WEB SITE Evan Eichler’s lab: http://genetics.gene.cwru.edu/eichler/ The positional cloning of a human disease gene is a landmark event. Good examples are provided by Bardet–Biedl syndrome (BBS), for which several genes have recently been identified (see Highlights, July). But these landmarks have now assumed greater significance, because some cases of BBS seem to require at least three mutant alleles. As a clear-cut example of ‘triallelic’ inheritance, BBS might therefore provide some general lessons about the interaction between genes, and its influence on phenotype. BBS is usually thought of as an autosomalrecessive disorder that is characterized by a broad and variable phenotype, involving retinal dystrophy, obesity, mental retardation, renal defects and polydactyly. BBS is also genetically heterogeneous — defects at several loci can cause the disease. So, when Katsanis et al. screened patients for mutations in two known BBS genes (BBS2 and BBS6), it was no surprise to find mutations in these genes in only some of the patients. What was more surprising was that some affected individuals carried three mutant alleles — for example, two in BBS2 and one in BBS6. The authors propose that BBS is inherited, at least in some individuals, in a triallelic fashion, although further BBS genes will need to be identified to determine the generality of this observation. In an accompanying Perspective, Arthur Burghes and colleagues, describe BBS inheritance as autosomal recessive with a modifier of penetrance, and liken BBS to other disorders in which modifier loci have been identified. However these unexpected findings are described, BBS provides a clear example of a phenotype that is determined by interactions between a small number of genes. BBS inheritance therefore lies somewhere between Mendelian and polygenic. As researchers move towards a more molecular understanding of BBS pathology, this disorder will become a valuable source of information about potential genetic interactions in more complex genetic disease. Mark Patterson References and links ORIGINAL RESEARCH PAPER Katsanis, N. et al. Triallelic inheritance in Bardet–Biedl syndrome, a Mendelian recessive disorder. Science 293, 2256–2259 (2001) FURTHER READING Burghes, A. H. M. et al. The land between Mendelian and multifactorial inheritance. Science 293, 2213–2214 (2001) | Nadeau, J. H. Modifier genes in mice and humans. Nature Rev. Genet. 2, 165–174 (2001) WEB SITES James Lupski’s lab: http://imgen.bcm.tmc.edu/molgen/lupski/ Bardet–Biedl syndrome links: http://www.kumc.edu/gec/support/laurmoon.html A bridge to complex disease H U M A N G E N E T I C S

A primate example of positive selection

www.nature.com/reviews/genetics Segmental duplications are a common feature of the human genome and they can be very difficult to sequence. But they should not be ignored. Through careful analysis of one set of segmental duplications, Matthew Johnson and colleagues have discovered a new human gene family. Furthermore, this gene family seems to have been subjected to powerful positive selection. The authors focused on a 20-kb repeated segment that is confined to a 15-Mb region of chromosome 16. There are 15 copies of this segment, which have very high levels of sequence similarity. To study the evolutionary history of the duplicated segments, the authors identified the orthologous sequences in a series of primates. Their analysis showed that the segment is only present in one or two copies in Old World monkeys, such as the baboon. By contrast, in great apes such as gorillas, which are more closely related to humans, there are 9–30 copies of the segment. Overall, the segment seems to have been duplicated recently and independently in several primate lineages, after the divergence of humans and great apes from Old World monkeys. When the sequence of the human segment was used in a database search, the authors found that an expressed sequence lurks within the repeated segment, although no homologues could be found in other organisms. Surprisingly, sequence comparisons of the human repeats showed that the putative protein-coding regions are five times more divergent compared with the non-coding regions of the repeat. This indicates that the gene might be under positive selection for adaptive mutations. In support of this, the ratio of nonsynomous to synonymous amino-acid changes was significantly greater than 1, and on the basis of comparisons with the primate sequences, evidence of positive selection could be found during the divergence of the great ape and human lineages. Despite the relatively recent duplication events, the gene has undergone major evolutionary change. Positive selection for amino-acid substitutions indicates an important function, which, for this gene family, is as yet unknown. Nevertheless, these observations attest to the importance of duplication and divergence as key evolutionary mechanisms. So, although segmental duplications can be a positive nuisance for genome sequencers, they provide some fascinating clues about the history of our genome. Mark Patterson References and links ORIGINAL RESEARCH PAPER Johnson, M. E. et al. Positive selection of a gene family during the emergence of humans and African apes. Nature 413 , 514–519 (2001) WEB SITE Evan Eichler’s lab: http://genetics.gene.cwru.edu/eichler/ The positional cloning of a human disease gene is a landmark event. Good examples are provided by Bardet–Biedl syndrome (BBS), for which several genes have recently been identified (see Highlights, July). But these landmarks have now assumed greater significance, because some cases of BBS seem to require at least three mutant alleles. As a clear-cut example of ‘triallelic’ inheritance, BBS might therefore provide some general lessons about the interaction between genes, and its influence on phenotype. BBS is usually thought of as an autosomalrecessive disorder that is characterized by a broad and variable phenotype, involving retinal dystrophy, obesity, mental retardation, renal defects and polydactyly. BBS is also genetically heterogeneous — defects at several loci can cause the disease. So, when Katsanis et al. screened patients for mutations in two known BBS genes (BBS2 and BBS6), it was no surprise to find mutations in these genes in only some of the patients. What was more surprising was that some affected individuals carried three mutant alleles — for example, two in BBS2 and one in BBS6. The authors propose that BBS is inherited, at least in some individuals, in a triallelic fashion, although further BBS genes will need to be identified to determine the generality of this observation. In an accompanying Perspective, Arthur Burghes and colleagues, describe BBS inheritance as autosomal recessive with a modifier of penetrance, and liken BBS to other disorders in which modifier loci have been identified. However these unexpected findings are described, BBS provides a clear example of a phenotype that is determined by interactions between a small number of genes. BBS inheritance therefore lies somewhere between Mendelian and polygenic. As researchers move towards a more molecular understanding of BBS pathology, this disorder will become a valuable source of information about potential genetic interactions in more complex genetic disease. Mark Patterson References and links ORIGINAL RESEARCH PAPER Katsanis, N. et al. Triallelic inheritance in Bardet–Biedl syndrome, a Mendelian recessive disorder. Science 293, 2256–2259 (2001) FURTHER READING Burghes, A. H. M. et al. The land between Mendelian and multifactorial inheritance. Science 293, 2213–2214 (2001) | Nadeau, J. H. Modifier genes in mice and humans. Nature Rev. Genet. 2, 165–174 (2001) WEB SITES James Lupski’s lab: http://imgen.bcm.tmc.edu/molgen/lupski/ Bardet–Biedl syndrome links: http://www.kumc.edu/gec/support/laurmoon.html A bridge to complex disease H U M A N G E N E T I C S

A bridge to complex disease: Human genetics

www.nature.com/reviews/genetics Segmental duplications are a common feature of the human genome and they can be very difficult to sequence. But they should not be ignored. Through careful analysis of one set of segmental duplications, Matthew Johnson and colleagues have discovered a new human gene family. Furthermore, this gene family seems to have been subjected to powerful positive selection. The authors focused on a 20-kb repeated segment that is confined to a 15-Mb region of chromosome 16. There are 15 copies of this segment, which have very high levels of sequence similarity. To study the evolutionary history of the duplicated segments, the authors identified the orthologous sequences in a series of primates. Their analysis showed that the segment is only present in one or two copies in Old World monkeys, such as the baboon. By contrast, in great apes such as gorillas, which are more closely related to humans, there are 9–30 copies of the segment. Overall, the segment seems to have been duplicated recently and independently in several primate lineages, after the divergence of humans and great apes from Old World monkeys. When the sequence of the human segment was used in a database search, the authors found that an expressed sequence lurks within the repeated segment, although no homologues could be found in other organisms. Surprisingly, sequence comparisons of the human repeats showed that the putative protein-coding regions are five times more divergent compared with the non-coding regions of the repeat. This indicates that the gene might be under positive selection for adaptive mutations. In support of this, the ratio of nonsynomous to synonymous amino-acid changes was significantly greater than 1, and on the basis of comparisons with the primate sequences, evidence of positive selection could be found during the divergence of the great ape and human lineages. Despite the relatively recent duplication events, the gene has undergone major evolutionary change. Positive selection for amino-acid substitutions indicates an important function, which, for this gene family, is as yet unknown. Nevertheless, these observations attest to the importance of duplication and divergence as key evolutionary mechanisms. So, although segmental duplications can be a positive nuisance for genome sequencers, they provide some fascinating clues about the history of our genome. Mark Patterson References and links ORIGINAL RESEARCH PAPER Johnson, M. E. et al. Positive selection of a gene family during the emergence of humans and African apes. Nature 413 , 514–519 (2001) WEB SITE Evan Eichler’s lab: http://genetics.gene.cwru.edu/eichler/ The positional cloning of a human disease gene is a landmark event. Good examples are provided by Bardet–Biedl syndrome (BBS), for which several genes have recently been identified (see Highlights, July). But these landmarks have now assumed greater significance, because some cases of BBS seem to require at least three mutant alleles. As a clear-cut example of ‘triallelic’ inheritance, BBS might therefore provide some general lessons about the interaction between genes, and its influence on phenotype. BBS is usually thought of as an autosomalrecessive disorder that is characterized by a broad and variable phenotype, involving retinal dystrophy, obesity, mental retardation, renal defects and polydactyly. BBS is also genetically heterogeneous — defects at several loci can cause the disease. So, when Katsanis et al. screened patients for mutations in two known BBS genes (BBS2 and BBS6), it was no surprise to find mutations in these genes in only some of the patients. What was more surprising was that some affected individuals carried three mutant alleles — for example, two in BBS2 and one in BBS6. The authors propose that BBS is inherited, at least in some individuals, in a triallelic fashion, although further BBS genes will need to be identified to determine the generality of this observation. In an accompanying Perspective, Arthur Burghes and colleagues, describe BBS inheritance as autosomal recessive with a modifier of penetrance, and liken BBS to other disorders in which modifier loci have been identified. However these unexpected findings are described, BBS provides a clear example of a phenotype that is determined by interactions between a small number of genes. BBS inheritance therefore lies somewhere between Mendelian and polygenic. As researchers move towards a more molecular understanding of BBS pathology, this disorder will become a valuable source of information about potential genetic interactions in more complex genetic disease. Mark Patterson References and links ORIGINAL RESEARCH PAPER Katsanis, N. et al. Triallelic inheritance in Bardet–Biedl syndrome, a Mendelian recessive disorder. Science 293, 2256–2259 (2001) FURTHER READING Burghes, A. H. M. et al. The land between Mendelian and multifactorial inheritance. Science 293, 2213–2214 (2001) | Nadeau, J. H. Modifier genes in mice and humans. Nature Rev. Genet. 2, 165–174 (2001) WEB SITES James Lupski’s lab: http://imgen.bcm.tmc.edu/molgen/lupski/ Bardet–Biedl syndrome links: http://www.kumc.edu/gec/support/laurmoon.html A bridge to complex disease H U M A N G E N E T I C S

Functional genomics. Comprehensive interference..

With the publication of its genome sequence, Arabidopsis now ranks among the elite of post-genomic organisms, alongside yeast, worms and flies. Being sessile organisms, it’s not surprising that plants have evolved their own molecular peculiarities — Arabidopsis lacks stars of some well-known signal transduction pathways, such as Wnt and Notch, but has many unique protein families, especially among transcription factors. The complete sequence, the quality of which surpasses all other whole-genome sequences published so far, will simplify forward-mutational analyses, although the frequent gene duplications raise the spectre of functional redundancy. Surveying gene expression patterns provides another avenue for investigating gene function, and a recent microarray analysis of gene expression during the Arabidopsis immune defence response illustrates this point.

Behavioural genetics: Watching the clock

VOLUME 2 | JULY 2001 | 487 The identification of the Drosophila period (per) locus — published in 1971 by Ronald Konopka and Seymour Benzer — was a singular landmark in behavioural genetics research. They were the first to show that genetics could be used to dissect behaviour, in this case the internal clock, or circadian rhythm, that underlies many aspects of behaviour in diverse organisms. Since then, many more genes involved in circadian rhythms have been identified (at least nine in mammals); the basic molecular mechanism of the clock has been defined; and the general features of the clock are known to be conserved in most living organisms. But little is known about the connections between the cellular workings of the clock and the patterns of behaviour that it influences. Two new studies provide valuable insights into how such knowledge might be obtained. The study by Zheng et al. begins with an analysis of the mouse gene Per (mPer1) — one of three mammalian per homologues. By knocking out Per, the authors show that its clock functions differ from those of Per2 (mPer2). This reinforces the idea that the core mechanism of the mammalian clock is more complex than in Drosophila. To investigate further the different roles of Per and Per2, the authors used microarray studies to identify genes that might be regulated by the clock in Per, Per2 and double-knockout mutants. They found 16 genes with reproducible circadian expression that was dependent on Per, Per2, or both genes. Two genes in particular showed circadian expression and were dependent on both Per and Per2: Alas1 and Alas2, which encode rate-limiting enzymes involved in haem biosynthesis. Because haem is an essential part of many proteins, such as metabolic and signalling proteins, the authors speculate that the circadian regulation of these genes might feed into a range of physiological processes. In the second study, Shimomura et al. used a quantitative genetics approach to identify loci that underlie the difference in circadian rhythms (as measured by wheel-running behaviour) between two inbred mouse strains. Their phenotypic analysis measured five aspects of circadian behaviour, such as the length of the circadian cycle and the phase angle of entrainment — how the onset of activity relates to the light–dark cycles. In addition to searching for independent quantitative trait loci, the authors also looked for epistatic interactions between loci, by incorporating a genome-wide search for loci that had an effect only in combination with another locus. They found 14 loci that had a significant effect on circadian behaviour, including several that interacted epistatically. Importantly, most of these loci mapped to regions outside those containing known mammalian circadian genes. In the 30 years since period was first identified, geneticists have built on the sound foundations laid by the pioneers of this field, but before the vision of Benzer and his colleagues can be realized, a good deal more work needs to be done. Both of these new lines of research hold great promise for finding new genes that function somewhere between the cellular clock and its output — circadian behaviour. Mark Patterson

The rat pack

NATURE REVIEWS | GENETICS VOLUME 2 | JANUARY 2001 | 7 The rat pack For decades, the rat has been used as a model system for studying human physiology and disease. In particular, there are excellent rat models for multifactorial diseases, the genetic components of which are being hunted down with increasing vigour. Consequently, there is strong motivation to develop genetic and genomic resources for the rat, so that the wealth of phenotypic and physiological data can be exploited in genetic analyses. The data and resources are accumulating rapidly — the rat genome is scheduled to be sequenced to at least fourfold coverage by the end of 2002 — and to provide centralized access to this information, the Rat Genome Database (RGD) was launched in June, 2000. RGD is the result of an international collaboration of rat researchers and is hosted at the Medical College of Wisconsin. The information available at RGD includes maps (genetic and physical), genes, ESTs, simple-sequence length polymorphisms and phenotypic data for 48 important inbred rat strains. Rat genes are linked to human and mouse homologues and to related information, such as NCBI’s LocusLink, the Ratmap database and the Rat Gene Index at The Institute for Genomic Research. RGD also provides tools for data analysis, such as Metagene. Users can submit genome sequence to Metagene and the sequence is analysed by seven popular gene-prediction algorithms. Results are aligned for all packages, allowing the user to compare the output and to assess the statistical significance of the predicted coding regions. And if you’re stuck for a rat person to talk to, the Rat Community Forum, also hosted by the Medical College of Wisconsin, is a good first port of call. Mark Patterson WEB WATCH