The human genome and the future of medicine

The completion of the draft sequence of the human genome in 2001 was one of the great milestones of science.1,2 However, this event is important more for what it has begun rather than for what it has finished. Determining the sequence has laid the foundations for determining the complete set of proteins that are produced in the human (the "proteome"), but we do not know the function of most of these proteins. Even when a protein sequence allows a reasonably confident prediction of the biochemical action of the protein, such as a probable tyrosine kinase or serine protease, the role of these proteins in human physiology and development has not yet been determined. This will take years to sort out, and will often rely on information about equivalent proteins in other organisms, both vertebrate and invertebrate.

Yet, even this is just the start of unravelling the meaning of the information stored in our genome. We still have to ascertain how variation in these proteins, and especially the sequences that regulate their production, affect our physiological networks and our individual characteristics, including susceptibility to disease.

It is important to appreciate that the Human Genome Project is just the flagship of a fleet of studies to explore the molecular and genetic basis of life and its diversity, which will provide the scientific and technological "scaffolding" for understanding the human genome and human biology. An updated and comprehensive list of completed and in-progress genome projects can be obtained from the Genomes OnLine Database,3 which also has excellent links to relevant publications and information.

The first complete genome sequence, that of Haemophilus influenzae, was published in 1995. By late 2002, the sequences of 125 organisms had been completed, including 110 prokaryotes and 15 eukaryotes (including mouse, puffer fish, fruit fly, mosquito, nematode, rice, mustard plant, malaria, and two yeasts). At least 580 other genome-sequencing projects (346 prokaryotes and 235 eukaryotes) are in progress, including those of the chimpanzee, cow, pig, dog, rat, chicken, toad, salmon, bee, tick, shrimp, corn, wheat, tomato, and eucalyptus. Current sequencing projects also include the genomes of many fungal, protozoan and invertebrate pathogens of humans and other mammals, including trypanosomes and helminths. Sooner rather than later, the genomes of all organisms and variants of scientific or practical interest will be sequenced. The information generated by these projects will allow us to understand the genetic programming and evolution of life in exquisite detail.

What we have learned so far from genome sequencing is how little we know about our genetic programming (Box 1). However, there have been some surprising observations. The first is that the number of protein-coding genes in humans is much lower than expected — it had been predicted that humans would have at least 100 000 genes (encoding different proteins), but this is not so. Humans have about 30 000 protein-coding genes, similar to other vertebrates,1,2,4,5 although the repertoire of protein isoforms is greatly expanded by alternative splicing of pre-mRNA.6 Nevertheless, only about 1.5% of our genome is occupied by protein-coding sequences, which raises the question of what function the rest of the genome has, and how the complex developmental program that unfolds following fertilisation is encoded within it.

The second surprise is the amazing conservation of protein-coding genes between different organisms. Recognisable equivalents of 99% of the proteins in humans can be found in the mouse, and vice versa, and about 95% are held in common.4 Many human proteins are also very similar in structure and function to (and in some cases interchangeable with) those found in invertebrates, and include proteins controlling eye development, body plan, and circadian rhythms.7,8 Moreover, of the about six million (0.1%) DNA-sequence differences between individual humans (in their diploid genomes), only about 20 000 are in protein-coding sequences, and most of these encode silent changes, which do not affect the amino acid sequence of the protein.2 Therefore, apart from mutations that render particular proteins non-functional (the cause of classical monogenic diseases like cystic fibrosis and thalassaemia), individual humans have much the same set of proteins, and humans and mice have a very similar set of proteins. Thus, differences in the proteins alone may not be sufficient to explain the differences between individuals and between species.

The third and greatest surprise is the enormous amount of non-protein-coding RNA that is transcribed, and the extensive sequence conservation between the human and mouse genomes in introns (intervening sequences that interrupt protein-coding sequences) and "intergenic" regions.4 These findings suggest that these regions have important, but not yet understood, functions. Introns comprise, on average, 95% of the primary sequence of protein-coding sequences in humans.1,2 In addition, it appears likely that non-protein-coding RNAs represent at least half and perhaps as much as three-quarters of all transcripts in humans.9,10 The transcriptional activity on human chromosomes 21 and 22 is at least an order of magnitude greater than expected from known protein-coding sequences,11 and around 50% of all transcripts in the mouse do not contain substantial protein-coding sequences.12 The expression of these non-coding RNAs is different in different cells, and a number of such RNAs have been implicated in diseases such as cancer, autism, spinocerebellar ataxia type 8, and cartilage–hair hypoplasia. Thus, we may have around 100 000 genes after all, but many, if not most, of them do not code for protein.9 Another way of looking at this is that, although only about 1.5% of the human genome encodes protein, more than 50% of the sequences in the genome are actually expressed in a developmentally regulated manner.10 Either the human genome is replete with useless transcription, or these RNAs are fulfilling unexpected functions. The strong possibility exists that this massive output of non-coding RNA sequences represents an internal communication mechanism to coordinate the complex suites of gene expression and programmed responses that underpin differentiation and development, and that this is a key part of the control architecture that enabled human complexity to evolve.9,10 That is, most of our genome may be specifying the plans for the assembly, not just the components (proteins), of the system.

Predicting the future is risky. Experience suggests that things will change faster than we expect, but usually not in the way that we expect. Some things expected to have happened quickly, such as the development of genetically engineered vaccines against malaria, have not yet eventuated, whereas other things have happened that we did not anticipate, such as the development of the polymerase chain reaction (PCR) in the late 1980s, which made DNA-based diagnostics a reality and revolutionised genetic engineering overnight.13 Although the field of human genomics is in its infancy, there are nevertheless a number of trends emerging in terms of its impact on medicine and health.

Lest the sceptics believe that some of these predictions are optimistic, consider the changes that occurred during the twentieth century. In 1901, no one knew that DNA contains our genetic inheritance, let alone dreamed that the complete sequence of the human genome would be deciphered in 2001. It has been said that "genetic prediction of individual risks of disease and responsiveness to drugs will reach the medical mainstream in the next decade or so. The development of designer drugs, based on a genomic approach to targeting molecular pathways that are disrupted in disease, will follow soon after".16 The pace of acquisition of knowledge is accelerating and so will the pace of change. These generally positive developments will also bring problems, some of which are being addressed locally by the current inquiry into the Protection of Human Genetic Information being conducted by the Australian Law Reform Commission and the Australian Health Ethics Committee,27 and a newly commissioned Inquiry into Gene Patenting.28

Underlying many of these particular issues are deeper ethical and philosophical tensions, as, on the one hand, some people are concerned about interfering with natural life processes ("playing God"), and, on the other hand, we all want to live longer and healthier lives. However, we should bear in mind that our ethical views evolve as quickly as the world around us, especially when we are faced with real choices rather than theoretical dangers. What will happen when we work out the genetic factors underpinning human intellectual and artistic potential is anybody's guess, especially if artificial intelligence has in the meantime become a reality through advances in computational science. To be around to find out would be nice.

2: Gene expression during exposure of human endothelial cells to hypoxia

Pictured is a subset of a 20 000-gene array, with the colour temperature indicative of the level of transcripts binding to each spot. Illustration courtesy of Dr Sean Grimmond, Institute for Molecular Bioscience, University of Queensland.