Discovering cutting-edge methods for the synthesis of chiral amines is essential since many pharmaceutical molecules contain this motif. Moreover, to meet the challenges of sustainability, it is imperative to develop green, chemo-, regio- and stereo-selective processes. Biocatalysis meets its expectations but the known enzymatic methods to obtain amines suffer from an equilibrium shift requirement, or the need to recycle cofactors. On the contrary, enzymes forming C-C bonds do not have these drawbacks and would allow, in a single thermodynamically favored and highly selective step, to build the carbon backbone of target molecules while providing access to the amino moiety. Moreover, they display a broad spectrum of electrophilic carbonyl substrates. Thus, we assume that they will also be able to accept imines. With this in mind, we will explore two families of enzymes: aldolases and synthases which should allow to catalyze a reaction between a nucleophile and an imine as electrophile. The 4-amino-3,4-dideoxy-D-arabino-heptulosonate 7-phosphate (ADAHP) synthases have already shown their ability to use an imine as an electrophile. However, this weakly documented activity has never been exploited in synthesis. We propose to perform the first large-scale exploration of the biodiversity of the ADAHP synthases family to evaluate their catalytic capacities for chiral amines synthesis. In the case of aldolases, whose electrophilic substrate is an aldehyde, the aim is to identify a promiscuous activity towards imines, on the basis of our recent studies which have demonstrated the great plasticity of their active site. This would reveal a new activity called "aldaminase" which will bring a major advance in the well-known asymmetric Mannich reaction by allowing the use of enolizable aldehydes. The enzymes will be searched in the genomic biodiversity thanks to a bioinformatic approach, to build a collection of several hundreds of synthases and aldolases with a special focus on extremophilic microorganisms and from metagenomes (Tara project). Molecular probes will be synthetized to perform a high throughput screening in the retroaldamination direction using spectrophotometric assays. Indeed, as aldaminase activity was never described within the aldolase family, and considering the problem of imine instability in water, we hypothesize that a first detection of the retro-Mannich reaction is a more achievable goal. On emerging hits, modifications will then be undertaken by rational mutagenesis approaches, with the support of molecular modeling, to both increase their activities towards imines and reduce their natural abilities to convert the corresponding carbonyl into aldol. The best biocatalysts will be tested in deep eutectic solvents, allowing to increase water sensitive imines concentration. In the presence of efficient "aldaminases", the synthesis of essential synthons for the preparation of various pharmaceutical products will be considered. This highly interdisciplinary project brings together two partners whose collaboration has already been fruitful. Partner 1 (ICCF) brings a strong expertise in the field of biocatalytic processes and particularly in the study and exploitation of carboligases. Physical chemists specialized in the study of eutectic solvents are also members of the partner’s 1 team. Partner 2 (CEA/Genoscope), brings a unique expertise in the fields of exploration of catalytic capacities in the biodiversity on a large scale, which combines bioinformatics methods for the analysis of (meta)genomes, cloning and high throughput screening. This project will allow the development of the largest collection of carboligases existing to date. The ALDAMINASE project promises remarkable innovations in the field of biocatalysis and should constitute a major advance in sustainable synthesis of active pharmaceutical ingredients.