Screening to assess brain function is the least satisfactory aspect of phenotyping, and yet it is the most crucial. Before the advent of optogenetics, each different neuroscience technique could cover only one brain function parameter: electrical or chemical, spatial or temporal, etc. Brain imaging cannot detect neurotransmitter activity; a cDNA microarray is only a snapshot of transcripts; only a small number of neurons can be covered by average evoked potentials. Neuroscientists are always endeavoring to deduce what is happening in the brain on the basis of partial data provided by a range of techniques. The recent advent of optogenetics has now made it possible to link spatial, temporal, and functional aspects of a nerve tissue.

Optogenetics falls within the scope of a top-down approach as several levels of integration of the organism can be managed together. Optogenetics is the “integration of optics and genetics to achieve gain or loss-of-function of well-defined events within specific cells of living tissues” [23]. Neuronal activity and gain/loss-of-function induced by genetic changes have different timescales. Action potential is the electrical result of the polarization-depolarization activity of the neuronal membrane, and is a rapid phenomenon, with a duration of around 1 ms and a conduction speed ranging from 7 to 120 meters per second. Exchanges at the synaptic level are also rapid. A dedicated approach is needed for the neuronal impairment required to model neurodegenerative disorders and channelopathies. Genetic changes have long-term effects. Microbial opsin genes are able to provide one response which is compatible with the speed of the synaptic processes, (1) because they bind to channel mechanisms (ion pump, channels) and (2) because they generate photo-inducible currents when introduced into neurons. Opsin genes, as is the case for other genes, can be driven towards a specific neuronal target until a specific reporter or viral vector becomes available. The approach is possible both in vitro (using brain slices) and in vivo. Haubensak et al. [24], who explored the neuronal control exercised by the amygdala on the fear conditioning response, used both approaches, combining optogenetics and a patch-clamp for brain slices and for free moving animals. With the possibility of combining fluorescent markers/reporters, protein changes can be observed in zebrafish, either during development or during pharmacological treatment. Most of the available strains express fluorescent proteins under certain conditions. The method does not have the accuracy of optogenetics, but it does not require a very complex setup and can often provide an overview before embarking on a full and expansive investigation.

ASD, as defined by the DSM and specifically by DSM-5, is a construct based on the identification of a common denominator in a number of diseases. The new criteria are (1) “persistent deficits in social communication and social interaction across multiple contexts” and (2) “restricted, repetitive patterns of behavior, interests, or activities.” The common denominator here is more common than for the autistic triad, but is nevertheless a construct embracing heterogeneous characteristics. Throughout the book we have seen that the selection of these criteria has made ASD more heterogeneous. Fifty-four percent of patients with autism have cognitive impairments; 50–80 % have sleep difficulties, 35 % present epileptic features, and more than 80 % have motor difficulties. Behind this clinical heterogeneity is etiological heterogeneity. It has been noted in the course of the book that autism is a collection of different disorders grouped together on the basis of common features. This has two major consequences for ASD expressed here as two caveats.

Caveat number 1. As ASD is a group of diseases, it is unreasonable to expect that a new drug or educational treatment be found as a general cure for ASD. Alleviation has been achieved and reported in the literature, but the impact is no doubt nonspecific. These approaches must be continued whenever a beneficial effect is observed. The most promising prospect has been with the screening of new compounds for rescuing functions impaired in the rare diseases associated with autism.

Caveat Number 2. There have been many discussions and arguments on standardizing the behavioral tests used to characterize organism models (in particular, mouse models), and the issue is even more difficult for ASD. Behavioral analyses must factor in typical core features (repetitive behavior and impaired sociability) for each of the ASD diseases, while also considering the specificity of the disease and any associated traits not in the DSM-5.