Automatic Prediction of Facial Trait Judgments: Appearance vs. Structural Models

Evaluating other individuals with respect to personality characteristics plays a crucial role in human relations and it is the focus of attention for research in diverse fields such as psychology and interactive computer systems. In psychology, face perception has been recognized as a key component of this evaluation system. Multiple studies suggest that observers use face information to infer personality characteristics. Interactive computer systems are trying to take advantage of these findings and apply them to increase the natural aspect of interaction and to improve the performance of interactive computer systems. Here, we experimentally test whether the automatic prediction of facial trait judgments (e.g. dominance) can be made by using the full appearance information of the face and whether a reduced representation of its structure is sufficient. We evaluate two separate approaches: a holistic representation model using the facial appearance information and a structural model constructed from the relations among facial salient points. State of the art machine learning methods are applied to a) derive a facial trait judgment model from training data and b) predict a facial trait value for any face. Furthermore, we address the issue of whether there are specific structural relations among facial points that predict perception of facial traits. Experimental results over a set of labeled data (9 different trait evaluations) and classification rules (4 rules) suggest that a) prediction of perception of facial traits is learnable by both holistic and structural approaches; b) the most reliable prediction of facial trait judgments is obtained by certain type of holistic descriptions of the face appearance; and c) for some traits such as attractiveness and extroversion, there are relationships between specific structural features and social perceptions.

Analysis of the mechanisms controlling the activity of flp1, a chromosomal passenger protein in the fission Schizosaccharomyces pombe

ABSTRACT In S. cerevisiae, the protein phosphatase Cdc14pwt is essential far mitotic exit through its contribution to reducing mitotic CDK activity. But Cdc14pwt also acts as a mare general temporal coordinator of mid and late mitotic events by controlling the partitioning of DNA, microtubule stability and cytokinesis. Cdc14pwt orthologs are well conserved from yeasts to humans, and sequence comparison revealed the presence of three domains, A, B and C, of which A and B form the catalytic domain. Cdc14pwt orthologs are regulated (in part) through cell cycle dependent changes in their localization. Some of them are thought to be kept inactive by sequestration in the nucleolus during interphase. This is the case for flp1pwt, the single identified Cdc14pwt ortholog in the fission yeast S. pombe. In early mitosis, flp1pwt leaves the nucleolus and localizes to the kinetochores, the contractile ring and the mitotic spindle, suggesting that it has multiple substrates and regulates many mitotic processes. flp1D cells show a high chromosome loss rate and septation defects, suggesting a role for flp1wt in the fidelity of chromosome transmission and cytokinesis. The aim of this study is to characterize the mechanisms underlying flp1pwt functions and the control of its activity. A structure-function analysis has revealed that the presence of both A and B domains is required for biological function and for proper flp1pwt mitotic localization. In contrast, the C domain of flp1pwt is responsible for its proper nucleolar localization in G2/interphase. My data suggest that dephosphorylation of substrates by flp1pwt is not necessary for any changes in localization of flp1pwt except that at the medial ring. In that particular case, the catalytic activity of flp1pwt is required for efficient localization, therefore revealing an additional level of regulation. All the functions of flp1pwt assayed to date require its catalytic activity, emphasizing the importance of further identification of its substrates. As described for other orthologs, the capability of selfinteraction and phosphorylation status might help to control flp1pwt activity. My data suggest that flp1pwt forms oligomers in vivo and that phosphorylation is not essential far localization changes of the protein. In addition, the hypophosphorylated form of flp1pwt might be specifically involved in the promotion of cytokinesis. The results of this study suggest that multiple modes of regulation including localization, selfassociation and phosphorylation allow a fine-tuning regulation of flp1pwt phosphatase activity, and more generally that of Cdc14pwt family of phosphatases. RESUME Chez la levure S. cerevisiae, la protéine phosphatase Cdc14pwt est essentielle pour la sortie de mitose du fait de sa contribution dans la réduction d'activité des CDK mitotiques. Comme elle contrôle également le partage de l'ADN, la stabilité des microtubules et la cytokinèse, Cdc14pwt est en fait considérée comme un coordinateur temporel général des évènements de milieu et de fin de mitose. Les orthologues de Cdc14pwt sont bien conservés, des levures jusqu'à l'espèce humaine. Des comparaisons de séquence ont révélé la présence de trois domaines A, B et C, les deux premiers constituant le domaine catalytique. Ils sont régulés (en partie) via des changements dans leur localisation, eux-mêmes dépendants du cycle cellulaire. Plusieurs de ces orthologues sont supposés inactivés par séquestration dans le nucléole en interphase, ce qui est le cas de flp1pwt le seul orthologue de Cdc14pwt identifié chez la levure fissipare S, pombe. En début de mitose, flp1pwt quitte le nucléole et localise au niveau des kinetochores, de l'anneau contractile d'actine et du fuseau mitotique, ce qui laisse supposer de multiples substrats et fonctions. Comme les cellules délétées pour le gène flp1wt présentent un taux élevé de perte de chromosome et des défauts de septation, flp1pwt semble jouer un rôle dans la fidélité de la transmission du matériel génétique et la cytokinèse. Le but de cette étude est de caractériser les mécanismes impliqués dans les fonctions assurées par flp1pwt d'une part, et dans le contrôle de son activité d'autre part. Une analyse structure-fonction a révélé que la présence simultanée des deux domaines A et B est requise pour la fonction biologique de flp1pwt et sa localisation correcte pendant la mitose. Par contre, le domaine C de flp1pwt confère une localisation nucléolaire adéquate en G2/interphase. Mes données suggèrent que la déphosphorylation de substrats par flp1pwt est dispensable pour sa localisation correcte excepté celle à l'anneau médian, qui requiert dans ce cas, l'activité catalytique de flp1pwt, révélant ainsi un niveau de régulation supplémentaire. Toutes les fonctions de flp1 pwt testées jusqu'à présent nécessitent également son activité catalytique, ce qui accentue l'importance de l'identification future de ses substrats. Comme cela a déjà été décrit pour d'autres orthologues, la capacité d'auto-intéraction et le niveau de phosphorylation pourraient contrôler l'activité de flp1pwt. En effet, mes données suggèrent que flp1pwt forme des oligomères in vivo et que la phosphorylation n'est pas essentielle pour les changements de localisation observés pour la protéine. De plus, la forme hypophosphorylée de flp1pwt pourrait être spécifiquement impliquée dans la promotion de la cytokinèse. De multiples modes de régulation incluant la localisation, l'auto-association et la phosphorylation semblent permettre un contrôle fin et subtil de l'activité de la phosphatase flp1pwt, et plus généralement celle des protéines de la famille de Cdc14pwt.

Midregional pro-atrial natriuretic peptide and procalcitonin improve survival prediction in VAP.

Ventilator-associated pneumonia (VAP) affects mortality, morbidity and cost of critical care. Reliable risk estimation might improve end-of-life decisions, resource allocation and outcome. Several scoring systems for survival prediction have been established and optimised over the last decades. Recently, new biomarkers have gained interest in the prognostic field. We assessed whether midregional pro-atrial natriuretic peptide (MR-proANP) and procalcitonin (PCT) improve the predictive value of the Simplified Acute Physiologic Score (SAPS) II and Sequential Related Organ Failure Assessment (SOFA) in VAP. Specified end-points of a prospective multinational trial including 101 patients with VAP were analysed. Death <28 days after VAP onset was the primary end-point. MR-proANP and PCT were elevated at the onset of VAP in nonsurvivors compared with survivors (p = 0.003 and p = 0.017, respectively) and their slope of decline differed significantly (p = 0.018 and p = 0.039, respectively). Patients with the highest MR-proANP quartile at VAP onset were at increased risk for death (log rank p = 0.013). In a logistic regression model, MR-proANP was identified as the best predictor of survival. Adding MR-proANP and PCT to SAPS II and SOFA improved their predictive properties (area under the curve 0.895 and 0.880). We conclude that the combination of two biomarkers, MR-proANP and PCT, improve survival prediction of clinical severity scores in VAP.

Protein-ligand binding free energy estimation using molecular mechanics and continuum electrostatics. Application to HIV-1 protease inhibitors.

A method is proposed for the estimation of absolute binding free energy of interaction between proteins and ligands. Conformational sampling of the protein-ligand complex is performed by molecular dynamics (MD) in vacuo and the solvent effect is calculated a posteriori by solving the Poisson or the Poisson-Boltzmann equation for selected frames of the trajectory. The binding free energy is written as a linear combination of the buried surface upon complexation, SASbur, the electrostatic interaction energy between the ligand and the protein, Eelec, and the difference of the solvation free energies of the complex and the isolated ligand and protein, deltaGsolv. The method uses the buried surface upon complexation to account for the non-polar contribution to the binding free energy because it is less sensitive to the details of the structure than the van der Waals interaction energy. The parameters of the method are developed for a training set of 16 HIV-1 protease-inhibitor complexes of known 3D structure. A correlation coefficient of 0.91 was obtained with an unsigned mean error of 0.8 kcal/mol. When applied to a set of 25 HIV-1 protease-inhibitor complexes of unknown 3D structures, the method provides a satisfactory correlation between the calculated binding free energy and the experimental pIC5o without reparametrization.

An evaluation of volume-based morphometry for prediction of mild cognitive impairment and Alzheimer's disease.

Voxel-based morphometry from conventional T1-weighted images has proved effective to quantify Alzheimer's disease (AD) related brain atrophy and to enable fairly accurate automated classification of AD patients, mild cognitive impaired patients (MCI) and elderly controls. Little is known, however, about the classification power of volume-based morphometry, where features of interest consist of a few brain structure volumes (e.g. hippocampi, lobes, ventricles) as opposed to hundreds of thousands of voxel-wise gray matter concentrations. In this work, we experimentally evaluate two distinct volume-based morphometry algorithms (FreeSurfer and an in-house algorithm called MorphoBox) for automatic disease classification on a standardized data set from the Alzheimer's Disease Neuroimaging Initiative. Results indicate that both algorithms achieve classification accuracy comparable to the conventional whole-brain voxel-based morphometry pipeline using SPM for AD vs elderly controls and MCI vs controls, and higher accuracy for classification of AD vs MCI and early vs late AD converters, thereby demonstrating the potential of volume-based morphometry to assist diagnosis of mild cognitive impairment and Alzheimer's disease.

Identification of human IKK-2 inhibitors of natural origin (Part II): in Silico prediction of IKK-2 inhibitors in natural extracts with known anti-inflammatory activity.

Human inhibitor NF-κB kinase 2 (hIKK-2) is the primary component responsible for activating NF-κB in response to various inflammatory stimuli. Thus, synthetic ATP-competitive inhibitors for hIKK-2 have been developed as anti-inflammatory compounds. We recently reported a virtual screening protocol (doi:10.1371/journal.pone.0016903) that is able to identify hIKK-2 inhibitors that are not structurally related to any known molecule that inhibits hIKK-2 and that have never been reported to have anti-inflammatory activity. In this study, a stricter version of this protocol was applied to an in-house database of 29,779 natural products annotated with their natural source. The search identified 274 molecules (isolated from 453 different natural extracts) predicted to inhibit hIKK-2. An exhaustive bibliographic search revealed that anti-inflammatory activity has been previously described for: (a) 36 out of these 453 extracts; and (b) 17 out of 30 virtual screening hits present in these 36 extracts. Only one of the remaining 13 hit molecules in these extracts shows chemical similarity with known synthetic hIKK-2 inhibitors. Therefore, it is plausible that a significant portion of the remaining 12 hit molecules are lead-hopping candidates for the development of new hIKK-2 inhibitors.

Experimental analysis and computer prediction of CTF/NFI transcription factor DNA binding sites.

Accurate prediction of transcription factor binding sites is needed to unravel the function and regulation of genes discovered in genome sequencing projects. To evaluate current computer prediction tools, we have begun a systematic study of the sequence-specific DNA-binding of a transcription factor belonging to the CTF/NFI family. Using a systematic collection of rationally designed oligonucleotides combined with an in vitro DNA binding assay, we found that the sequence specificity of this protein cannot be represented by a simple consensus sequence or weight matrix. For instance, CTF/NFI uses a flexible DNA binding mode that allows for variations of the binding site length. From the experimental data, we derived a novel prediction method using a generalised profile as a binding site predictor. Experimental evaluation of the generalised profile indicated that it accurately predicts the binding affinity of the transcription factor to natural or synthetic DNA sequences. Furthermore, the in vitro measured binding affinities of a subset of oligonucleotides were found to correlate with their transcriptional activities in transfected cells. The combined computational-experimental approach exemplified in this work thus resulted in an accurate prediction method for CTF/NFI binding sites potentially functioning as regulatory regions in vivo.

Prediction of free imatinib concentrations based on total plasma concentrations in patients with gastrointestinal stromal tumours.

AIM: Total imatinib concentrations are currently measured for the therapeutic drug monitoring of imatinib, whereas only free drug equilibrates with cells for pharmacological action. Due to technical and cost limitations, routine measurement of free concentrations is generally not performed. In this study, free and total imatinib concentrations were measured to establish a model allowing the confident prediction of imatinib free concentrations based on total concentrations and plasma proteins measurements. METHODS: One hundred and fifty total and free plasma concentrations of imatinib were measured in 49 patients with gastrointestinal stromal tumours. A population pharmacokinetic model was built up to characterize mean total and free concentrations with inter-patient and intrapatient variability, while taking into account α1 -acid glycoprotein (AGP) and human serum albumin (HSA) concentrations, in addition to other demographic and environmental covariates. RESULTS: A one compartment model with first order absorption was used to characterize total and free imatinib concentrations. Only AGP influenced imatinib total clearance. Imatinib free concentrations were best predicted using a non-linear binding model to AGP, with a dissociation constant Kd of 319 ng ml(-1) , assuming a 1:1 molar binding ratio. The addition of HSA in the equation did not improve the prediction of imatinib unbound concentrations. CONCLUSION: Although free concentration monitoring is probably more appropriate than total concentrations, it requires an additional ultrafiltration step and sensitive analytical technology, not always available in clinical laboratories. The model proposed might represent a convenient approach to estimate imatinib free concentrations. However, therapeutic ranges for free imatinib concentrations remain to be established before it enters into routine practice.

Normal variation in fronto-occipital circuitry and cerebellar structure with an autism-associated polymorphism of CNTNAP2.

Recent genetic studies have implicated a number of candidate genes in the pathogenesis of Autism Spectrum Disorder (ASD). Polymorphisms of CNTNAP2 (contactin-associated like protein-2), a member of the neurexin family, have already been implicated as a susceptibility gene for autism by at least 3 separate studies. We investigated variation in white and grey matter morphology using structural MRI and diffusion tensor imaging. We compared volumetric differences in white and grey matter and fractional anisotropy values in control subjects characterised by genotype at rs7794745, a single nucleotide polymorphism in CNTNAP2. Homozygotes for the risk allele showed significant reductions in grey and white matter volume and fractional anisotropy in several regions that have already been implicated in ASD, including the cerebellum, fusiform gyrus, occipital and frontal cortices. Male homozygotes for the risk alleles showed greater reductions in grey matter in the right frontal pole and in FA in the right rostral fronto-occipital fasciculus compared to their female counterparts who showed greater reductions in FA of the anterior thalamic radiation. Thus a risk allele for autism results in significant cerebral morphological variation, despite the absence of overt symptoms or behavioural abnormalities. The results are consistent with accumulating evidence of CNTNAP2's function in neuronal development. The finding suggests the possibility that the heterogeneous manifestations of ASD can be aetiologically characterised into distinct subtypes through genetic-morphological analysis.

The evolution of the thyroid hormone distributor protein transthyretin in the order insectivora, class mammalia.

Thyroid hormones are involved in the regulation of growth and metabolism in all vertebrates. Transthyretin is one of the extracellular proteins with high affinity for thyroid hormones which determine the partitioning of these hormones between extracellular compartments and intracellular lipids. During vertebrate evolution, both the tissue pattern of expression and the structure of the gene for transthyretin underwent characteristic changes. The purpose of this study was to characterize the position of Insectivora in the evolution of transthyretin in eutherians, a subclass of Mammalia. Transthyretin was identified by thyroxine binding and Western analysis in the blood of adult shrews, hedgehogs, and moles. Transthyretin is synthesized in the liver and secreted into the bloodstream, similar to the situation for other adult eutherians, birds, and diprotodont marsupials, but different from that for adult fish, amphibians, reptiles, monotremes, and Australian polyprotodont marsupials. For the characterization of the structure of the gene and the processing of mRNA for transthyretin, cDNA libraries were prepared from RNA from hedgehog and shrew livers, and full-length cDNA clones were isolated and sequenced. Sections of genomic DNA in the regions coding for the splice sites between exons 1 and 2 were synthesized by polymerase chain reaction and sequenced. The location of splicing was deduced from comparison of genomic with cDNA nucleotide sequences. Changes in the nucleotide sequence of the transthyretin gene during evolution are most pronounced in the region coding for the N-terminal region of the protein. Both the derived overall amino sequences and the N-terminal regions of the transthyretins in Insectivora were found to be very similar to those in other eutherians but differed from those found in marsupials, birds, reptiles, amphibians, and fish. Also, the pattern of transthyretin precursor mRNA splicing in Insectivora was more similar to that in other eutherians than to that in marsupials, reptiles, and birds. Thus, in contrast to the marsupials, with a different pattern of transthyretin gene expression in the evolutionarily "older" polyprotodonts compared with the evolutionarily "younger" diprotodonts, no separate lineages of transthyretin evolution could be identified in eutherians. We conclude that transthyretin gene expression in the liver of adult eutherians probably appeared before the branching of the lineages leading to modern eutherian species.

Synthesis and immunological characterization of 104-mer and 102-mer peptides corresponding to the N- and C-terminal regions of the Plasmodium falciparum CS protein.

We investigated the immunogenicity and the conformational properties of the non-repetitive sequences of the Plasmodium falciparum circumsporozoite (CS) protein. Two polypeptides of 104 and 102 amino acids long, covering, respectively, the N- and C-terminal regions of the CS protein, were synthesized using solid phase Fmoc chemistry. The crude polypeptides were purified by a combination of size exclusion chromatography and RP-HPLC. Sera of mice immunized with the free polypeptides emulsified in incomplete Freund's adjuvant strongly reacted with the synthetic polypeptides as well as with native CS protein as judged by ELISA and IFAT assays. Most importantly, these antisera inhibited the sporozoite invasion of hepatoma cells. In addition, sera derived from donors living in a malaria endemic area recognized the CS 104- and 102-mers. Conformational studies of the CS polypeptides were also performed by circular dichroism spectroscopy showing the presence of a weakly ordered structure that can be increased by addition of trifluoroethanol. The obtained results indicate that the synthetic CS polypeptides and the natural CS protein share some common antigenic determinants and probably have similar conformation. The approach used in this study might be useful for the development of a synthetic malaria vaccine.

Exclusion of pulmonary embolism using C-reactive protein and D-dimer. A prospective comparison.

Our goal was to evaluate the diagnostic utility of C-reactive protein (CRP) alone or combined with clinical probability assessment in patients with suspected pulmonary embolism (PE), and to compare its performance to a D-dimer assay. We conducted a prospective study in which we performed a common immuno-turbidimetric CRP test and a rapid enzyme-linked immunosorbent assay (ELISA) D-dimer test in 259 consecutive outpatients with suspected PE at the emergency department of a teaching hospital. We assessed clinical probability of PE by a validated prediction rule overridden by clinical judgment. Patients with D-dimer levels > or = 500 microg/l underwent a work-up consisting of lower-limb venous ultrasound, spiral computerized tomography, ventilation-perfusion scan, or pulmonary angiography. Patients were followed up for three months. Seventy-seven (30%) of the patients had PE. The CRP alone had a sensitivity of 84% (95% confidence interval [CI).: 74 to 92%) and a negative predictive value (NPV) of 87% (95% CI: 78 to 93%) at a cutpoint of 5 mg/l. Overall, 61 (24%) patients with a low clinical probability of PE had a CRP < 5 mg/l. Due to the low prevalence of PE (9%) in this subgroup, the NPV increased to 97% (95% CI: 89 to 100%). The D-dimer (cutpoint 500 micro g/l) showed a sensitivity of 100% (95% CI: 95 to 100%) and a NPV of 100% (95% CI: 94 to 100%) irrespective of clinical probability and accurately rule out PE in 56 (22%) patients. Standard CRP tests alone or combined with clinical probability assessment cannot safely exclude PE.

Structural determinants involved in the activation and regulation of G protein-coupled receptors: lessons from the alpha1-adrenegic receptor subtypes.

The aim of a large number of studies on G protein-coupled receptors was centered on understanding the structural basis of their main functional properties. Here, we will briefly review the results obtained on the alpha1-adrenergic receptor subtypes belonging to the rhodopsin-like family of receptors. These findings contribute, on the one hand, to further understand the molecular basis of adrenergic transmission and, on the other, to provide some generalities on the structure-functional relationship of G protein-coupled receptors.

Genetic and karyotypic structure of a hybrid zone between the chromosomal races Cordon and Valais in the common shrew, Sorex araneus

Hybrid zones between genetically differentiated populations provide material to study evolutionary processes. Since the discovery of chromosomal races in Sorex araneus, contact zones have attracted attention of scientists. So far, studies on genetic markers in Sorex hybrid zones are missing. The acrocentric chromosomal race Cordon and the highly metacentric race Valais meet and hybridize at Les Houches in the Western Alps. On a transect through the hybrid zone, 273 shrews were caught at 15 localities over 4 years. Karyotype as well as the nuclear protein loci Alb and Pg were analyzed. F-st and F-is values were calculated by F-statistics. An analysis on pooled samples revealed the genetical differences between the hybridizing races as the only cause of population structuring. Genetical markers show dines with very strong frequency shifts at a mountain torrent, but behave differently through the zone. The performance of the torrent in maintaining the hybrid zone, selection against hybrids, possible assortative mating and linkage of the Valais Pg allele to a diagnostic chromosome arm, are discussed.

EADock : design of a new molecular docking algorithm and some of its applications

3 Summary 3. 1 English The pharmaceutical industry has been facing several challenges during the last years, and the optimization of their drug discovery pipeline is believed to be the only viable solution. High-throughput techniques do participate actively to this optimization, especially when complemented by computational approaches aiming at rationalizing the enormous amount of information that they can produce. In siiico techniques, such as virtual screening or rational drug design, are now routinely used to guide drug discovery. Both heavily rely on the prediction of the molecular interaction (docking) occurring between drug-like molecules and a therapeutically relevant target. Several softwares are available to this end, but despite the very promising picture drawn in most benchmarks, they still hold several hidden weaknesses. As pointed out in several recent reviews, the docking problem is far from being solved, and there is now a need for methods able to identify binding modes with a high accuracy, which is essential to reliably compute the binding free energy of the ligand. This quantity is directly linked to its affinity and can be related to its biological activity. Accurate docking algorithms are thus critical for both the discovery and the rational optimization of new drugs. In this thesis, a new docking software aiming at this goal is presented, EADock. It uses a hybrid evolutionary algorithm with two fitness functions, in combination with a sophisticated management of the diversity. EADock is interfaced with .the CHARMM package for energy calculations and coordinate handling. A validation was carried out on 37 crystallized protein-ligand complexes featuring 11 different proteins. The search space was defined as a sphere of 15 R around the center of mass of the ligand position in the crystal structure, and conversely to other benchmarks, our algorithms was fed with optimized ligand positions up to 10 A root mean square deviation 2MSD) from the crystal structure. This validation illustrates the efficiency of our sampling heuristic, as correct binding modes, defined by a RMSD to the crystal structure lower than 2 A, were identified and ranked first for 68% of the complexes. The success rate increases to 78% when considering the five best-ranked clusters, and 92% when all clusters present in the last generation are taken into account. Most failures in this benchmark could be explained by the presence of crystal contacts in the experimental structure. EADock has been used to understand molecular interactions involved in the regulation of the Na,K ATPase, and in the activation of the nuclear hormone peroxisome proliferatoractivated receptors a (PPARa). It also helped to understand the action of common pollutants (phthalates) on PPARy, and the impact of biotransformations of the anticancer drug Imatinib (Gleevec®) on its binding mode to the Bcr-Abl tyrosine kinase. Finally, a fragment-based rational drug design approach using EADock was developed, and led to the successful design of new peptidic ligands for the a5ß1 integrin, and for the human PPARa. In both cases, the designed peptides presented activities comparable to that of well-established ligands such as the anticancer drug Cilengitide and Wy14,643, respectively. 3.2 French Les récentes difficultés de l'industrie pharmaceutique ne semblent pouvoir se résoudre que par l'optimisation de leur processus de développement de médicaments. Cette dernière implique de plus en plus. de techniques dites "haut-débit", particulièrement efficaces lorsqu'elles sont couplées aux outils informatiques permettant de gérer la masse de données produite. Désormais, les approches in silico telles que le criblage virtuel ou la conception rationnelle de nouvelles molécules sont utilisées couramment. Toutes deux reposent sur la capacité à prédire les détails de l'interaction moléculaire entre une molécule ressemblant à un principe actif (PA) et une protéine cible ayant un intérêt thérapeutique. Les comparatifs de logiciels s'attaquant à cette prédiction sont flatteurs, mais plusieurs problèmes subsistent. La littérature récente tend à remettre en cause leur fiabilité, affirmant l'émergence .d'un besoin pour des approches plus précises du mode d'interaction. Cette précision est essentielle au calcul de l'énergie libre de liaison, qui est directement liée à l'affinité du PA potentiel pour la protéine cible, et indirectement liée à son activité biologique. Une prédiction précise est d'une importance toute particulière pour la découverte et l'optimisation de nouvelles molécules actives. Cette thèse présente un nouveau logiciel, EADock, mettant en avant une telle précision. Cet algorithme évolutionnaire hybride utilise deux pressions de sélections, combinées à une gestion de la diversité sophistiquée. EADock repose sur CHARMM pour les calculs d'énergie et la gestion des coordonnées atomiques. Sa validation a été effectuée sur 37 complexes protéine-ligand cristallisés, incluant 11 protéines différentes. L'espace de recherche a été étendu à une sphère de 151 de rayon autour du centre de masse du ligand cristallisé, et contrairement aux comparatifs habituels, l'algorithme est parti de solutions optimisées présentant un RMSD jusqu'à 10 R par rapport à la structure cristalline. Cette validation a permis de mettre en évidence l'efficacité de notre heuristique de recherche car des modes d'interactions présentant un RMSD inférieur à 2 R par rapport à la structure cristalline ont été classés premier pour 68% des complexes. Lorsque les cinq meilleures solutions sont prises en compte, le taux de succès grimpe à 78%, et 92% lorsque la totalité de la dernière génération est prise en compte. La plupart des erreurs de prédiction sont imputables à la présence de contacts cristallins. Depuis, EADock a été utilisé pour comprendre les mécanismes moléculaires impliqués dans la régulation de la Na,K ATPase et dans l'activation du peroxisome proliferatoractivated receptor a (PPARa). Il a également permis de décrire l'interaction de polluants couramment rencontrés sur PPARy, ainsi que l'influence de la métabolisation de l'Imatinib (PA anticancéreux) sur la fixation à la kinase Bcr-Abl. Une approche basée sur la prédiction des interactions de fragments moléculaires avec protéine cible est également proposée. Elle a permis la découverte de nouveaux ligands peptidiques de PPARa et de l'intégrine a5ß1. Dans les deux cas, l'activité de ces nouveaux peptides est comparable à celles de ligands bien établis, comme le Wy14,643 pour le premier, et le Cilengitide (PA anticancéreux) pour la seconde.