Why conventional methods come too late. Using conventional methods of diagnosis such as serum and urine analysis, early detection is not possible. These methods detect the consequences of infection — altered metabolites, altered cell counts, the body's response — and those appear only once the disease process is well advanced. As the chapter puts it, the presence of a pathogen is normally suspected only when it has produced a disease symptom, and by that time its concentration is already very high in the body.

How PCR gets there first. PCR detects the pathogen itself, by amplification of its nucleic acid. Because every cycle — denaturation, primer annealing, extension — doubles the number of copies of the target sequence, repeated cycles amplify the segment approximately a billion times. So a very low concentration of a bacterium or virus, at a time when the symptoms of the disease are not yet visible, becomes easily detectable.

Why specificity comes free. Amplification only occurs if the primer sites are present, and the primers are chosen to match sequences unique to the pathogen. A product therefore means that pathogen is present — it is a detection and an identification in one step.

Where it is used. PCR is now routinely used to detect HIV in suspected AIDS patients, to detect mutations in genes in suspected cancer patients, and to identify many other genetic disorders.

Why there is no signal. A single-stranded DNA or RNA tagged with a radioactive molecule — the probe — is allowed to hybridise to its complementary DNA in a clone of cells, and is then detected by autoradiography. Hybridisation depends on complementary base pairing. The probe is made complementary to the normal sequence, so where the gene is mutated the probe will not have complementarity with it, cannot bind, is washed away, and leaves no radioactivity on the film. The clone with the mutated gene therefore does not appear.

What the logic amounts to. The absence of a signal is the positive diagnosis. This is unusual and worth remembering, since in most assays a signal means a finding.

How it could mislead. A false positive diagnosis of mutation would follow from anything that stops hybridisation for a non-genetic reason — too little DNA in that clone, a failed transfer to the membrane, a degraded or badly labelled probe, or washing conditions too stringent. A false negative would follow if the mutation were very small: a change of a single base may still allow the probe to bind well enough to give a signal, so the mutated clone would appear on the film like a normal one. This is why a probe result is confirmed by PCR or sequencing.

Why milk. Medicines containing biological products are often expensive to make, and the expense is largely in scale and in purification. A dairy cow already secretes many litres a day of a protein-rich fluid, continuously, for years, and the fluid is collected non-invasively without harming the animal. The protein arrives already outside the cells, in a relatively simple mixture, which makes recovery far easier than lysing cells and separating a product from the whole cytoplasm.

Why a mammal rather than a bacterium. A mammalian cell can carry out processing that a bacterium cannot — correct folding, formation of disulphide bonds, cleavage of pro-forms and the addition of sugar groups. Recall the insulin problem: the main challenge with rDNA insulin was getting it assembled into a mature form, which was solved only by making chains A and B separately in E. coli and joining them chemically. For a larger or a glycosylated human protein, that workaround is not available. A cow makes the protein the way a human cell would.

The additional benefit in this case. Rosie's milk contained human alpha-lactalbumin and was nutritionally a more balanced product for human babies than natural cow-milk — so here the modified product was itself the useful thing, not merely a source to purify from.

Other examples of the same approach. The human protein α-1-antitrypsin, used to treat emphysema, and similar attempts for phenylketonuria and cystic fibrosis.

The biological argument. Going beyond the morality of such issues, genetic modification of organisms can have unpredictable results when such organisms are introduced into the ecosystem. A GM organism is not a chemical that degrades; it is a living thing that reproduces, spreads, interacts with other species and can pass its genes on. Consequences can therefore appear far from the field and long after release, which is exactly the situation in which laboratory testing alone is insufficient.

What the GEAC does about it. The Indian Government set up the Genetic Engineering Approval Committee, which makes decisions regarding the validity of GM research and the safety of introducing GM organisms for public services. Note the two distinct functions: judging the science, and judging the release. A body that can withhold approval for release is the only practical safeguard against an outcome that cannot be reversed once it has happened.

Why the moral and the biological cannot be separated here. Some ethical standards are required to evaluate the morality of all human activities that might help or harm living organisms — and whether an activity harms living organisms is a biological question. The two kinds of judgement are the same judgement approached from two sides.

A concrete illustration from this chapter. Bt crops reduce reliance on chemical pesticides, which is an environmental gain. But a pest population under continuous selection can evolve resistance, and a toxin expressed in every tissue of a widely grown crop may affect insects that were never the target. Neither the benefit nor the risk is a moral intuition; both are empirical, and both need assessing before release.

Definition. Biopiracy is the use of bio-resources by multinational companies and other organisations without proper authorisation from the countries and people concerned, and without compensatory payment.

The asymmetry. Most of the industrialised nations are rich financially but poor in biodiversity and traditional knowledge. In contrast, the developing and underdeveloped world is rich in biodiversity and traditional knowledge related to bio-resources. So one group holds the raw material and the accumulated knowledge, and the other holds the capital, the laboratories and access to the patent system. Traditional knowledge related to bio-resources can be exploited to develop modern applications, and can also be used to save time, effort and expenditure during commercialisation — which is precisely what makes it valuable to take.

Why the patent system compounds it. Certain companies are being granted patents for products and technologies that make use of genetic materials, plants and other biological resources long identified, developed and used by farmers and indigenous people of a specific region. The 1997 Basmati patent is the standard case: a ‘new’ variety derived from Indian farmers' varieties crossed with semi-dwarf varieties and claimed as an invention, with the patent extending to functional equivalents, so that other people selling Basmati rice could be restricted by it. Similar attempts were made over turmeric and neem.

What is being done. There has been growing realisation of the injustice, inadequate compensation and unequal benefit sharing between developed and developing countries, and some nations are developing laws to prevent unauthorised exploitation of their bio-resources and traditional knowledge. The Indian Parliament has cleared the second amendment of the Indian Patents Bill, which takes such issues into consideration, including patent terms of emergency provisions, research, and development initiatives. The chapter's own warning stands: if we are not vigilant and do not immediately counter these patent applications, others may encash on our rich legacy.