The December 28 1966 livestock breeding Act organized animal selection on a national level, creating organisations in charge of the different missions to be fufilled (the breeding establishment (EDE) for the identification and performances recording, artificial insemination centers, breeding associations or UPRA) and initiating a National Genetic Improvement Board (CNAG). This board is responsible for advising the Ministry on animal selection, and for defining the missions of the technical institutes. INRA received the "complementary" mission (complementary to its primary research activities) of managing the national zootechnical databases, evaluating breeding animals in ruminant species and pig and, more generally, contributing to the co-management of the French animal breeding system within CNAG. This Act is responsible for the philosophy found in the Animal Genetics division, based on the complementarity of the analysis and management of genetic variability. It is also a continuum between research and its applications in genetic improvement.
As early as the 1970s, the missions conferred to INRA by the Act oriented its research: genetic evaluation and selection procedures but also analysis of genetic determinism of phenotypic traits. Most of todays experimental layout was created at that time.
Research and applications are often closely related. INRA's inventive and charismatic leaders strongly contributed to the development of selection programs for French animal populations. Individual or progeny testing stations were developed for all species, sometimes at INRA and have become major tools both for selection and research. INRA, on its own experimental farms, compared genotypes of various origins, both French and others, such as the Vachotron at Bourges or the Pin-au-Haras, before any advice of dissemination. Sometimes, amongst the numerous animal lines selected for research, several have been disseminated and have had a brillant career (Vedette chicken, the INRA95 double-muscled bovine line for terminal cross breeds, hyperprolific porcine line, INRA401 sheep line now known as Romane breed).
INRA was a pioneer in informatics due to its implication in animal selection. In 1970, the National Genetic Data Processing Centre (CTIG) was created, in charge of the integrity and perenity of breeding data and of genetic evaluation. Initially, the databases only concerned a few selected traits. However, gradually, they were enriched with new information, specifically collected for selection or by-products of non-genetic applications benefiting largely from the advances in computing and automatisation. From the start, these databases were designed for three purposes: genetic evaluation, as support for the breeding industry, and research. The diversification of the data progressively led to an increase in the number of traits evaluated and the definition of new selection goals.
With this data (both phenotypic and genealogical), scientists performed genetic evaluations, i.e. estimations of the genetic value of breeding animals (EBV) from their performances and those of their relatives. Genetic evaluation is a basic tool for the breeder and the EBV determine in great part whether a potential breeding individual will be selected or disseminated, and its commercial value. INRA has a double legitimacy for performing genetic evaluations: at INRA, both methodological and technical expertise needed to perform such evaluations are available and INRA has a neutral role vis a vis the breeders. Genetic evaluation benefits much from statistical and computing progress. Breaking down a phenotype into a genetic effect and an environmental effect depends mainly on the development of an appropriate modeling and resolving a system of equations that are often complex. The development of computing tools, mainly since the beginning of the 1990s, has opened up new perspectives in terms of more refined evaluations: more traits, more complex statistical models, larger populations (sometimes even international), and more precise genetic determinisms. As soon as the CTIG was founded, scientists have been involved in all the steps of genetic evaluation, from theoretical developments to calculations via software development, working with thoroughness and a sense towards service that has been validated since 2006 by the ISO9001 certification for the activities of CTIG and the bovine evaluation.
Genetic improvement is a slow process, requiring a strong initial investment and support from farmers. When a selection program is started, it must be easy to implement and aimed at improving a trait that is heritable, measurable and low in cost. Indeed, financial means are limited, the initial population is only rarely competitive and the program is costly. By focussing on one main trait, the cost of performance testing is limited and that of eliminating animals as well. Then progressively, the breeders like the users see the interest in genetic progress and become more demanding, requesting in addition to the improvement of a single limiting trait, an increasing number of aptitudes. Some traits have a small heritibility and are difficult to measure, making the selection more costly and complex but also more balanced. Some correlated unfavourable responses must also be corrected. This need to widen the program is also at the origin of the enrichment of databases and the complexity of selection objectives. This strategy for a progressive development has been developed over the past fourty years.
Therefore, selection objectives have undergone an important evolution i.e. in the 1970s they aimed only at improving productivity, in the 1980s they integrated product quality and in the 1990s, functional traits (fertility, longevity, disease resistance). This tendency has continued in the 2000s, with explicit demands pertaining to the sensorial or nutritional quality of products, adaptation to the environment, behaviour and animal welfare, or environmental impact. Often, selection objectives have been ahead of economic constraints, by considering some traits for which the breeder is not paid.
The geneticists need to be prospective since selection is a process with much inertia, in particular in ruminants. Several years are often necessary to develop a method that measures a trait on a large scale and this must be considered first before selection. Once the information is available, 10 years are necessary in cattle between the time a selection objective is defined and the time that its results reach the breeder or the consumer.
Breeds creation and maintenance contribute to a large source of genetic diversity. During the last 40 years, genetic diversity has decreased due to selection; however today's ruminant diversity is much higher in France than in most other developed countries. This is a direct result of the breeding Act, which, through the mutualisation of many tools and the help provided to poorly profitable structures, has permitted to maintain a significant number of populations despite the high competititon of the most profitable populations in the economic conditions of a particular moment. Indeed, without the breeding Act, the countryside would be much less colourful. Of course during the last 40 years, the relationship between INRA and its partners has evolved. Initially, the other organisms did not have sufficient qualified staff and INRA was the main reference. Gradually, the technical institutes then the selection organisations have grown, recruting more technically competent staff. This has led to the inflexion of INRA's missions and changes in relationships. This may be illustrated by several examples: the development at the beginning of the 1990s of "evaluation groups", i.e. entities for dialogue with evaluation users; the renovation of the Genetic Information Systems that was initiated during the 90s by INRA was rapidly led collectively, even though the role of the CTIG at the centre of the network was not questionned; the development of Agenae illustrates perfectly the recent evolution with the industry being directly involed in the management and financing of research that interests the breeders, even if the applications are not immediate. This change which has accelerated during the last 15 years, varies with the economics of the industry but globally evolves. Therefore, the direct implication of INRA in selection tends to decrease progressively. Scientists are concentrating more on their explicit missions: research and innovation of course, but genetic evaluation too, with a long-term vision. In November 2018, with the application of the European Zootechnical Regulation, INRA is still in charge of the National Database but transferred its genetic evaluation activity to its industrial partners, reorienting its activity to research and development for future innovations in selection.
Biochemical genetics also developed early to study milk and blood proteins and the major histocompatiblity complex. Even though this research was limited by the technical tools available, it has produced many results and rapidly led to important applications. As early as the 1960s, parentage testing with blood groups was used by breeders. A major evolution of the field has been the explosion of molecular genetics beginning in the 1980s. Research progress is rapid and the potential applications are more numerous than before. To guarantee a rapid but harmonious development, INRA and its professional partners founded the GIE LABOGENA in 1994. The activity increased considerably and the offer was renewed: testing was extended to most species, along with traceability, parentage assignment, genotyping of important genes or those responsible for diseases. For INRA, until 2013, year of its privatisation, LABOGENA was a major tool for the rapid dissemination of its innovations and not just a genotyping platform. It is also one of the tools that linked INRA to its industrial partners in all species.