Intégration de méthodes de data mining dans le renseignement criminel : analyse par des structures issues de la théorie des graphes dans le profilage des stupéfiants

Data mining can be defined as the extraction of previously unknown and potentially useful information from large datasets. The main principle is to devise computer programs that run through databases and automatically seek deterministic patterns. It is applied in different fields of application, e.g., remote sensing, biometry, speech recognition, but has seldom been applied to forensic case data. The intrinsic difficulty related to the use of such data lies in its heterogeneity, which comes from the many different sources of information. The aim of this study is to highlight potential uses of pattern recognition that would provide relevant results from a criminal intelligence point of view. The role of data mining within a global crime analysis methodology is to detect all types of structures in a dataset. Once filtered and interpreted, those structures can point to previously unseen criminal activities. The interpretation of patterns for intelligence purposes is the final stage of the process. It allows the researcher to validate the whole methodology and to refine each step if necessary. An application to cutting agents found in illicit drug seizures was performed. A combinatorial approach was done, using the presence and the absence of products. Methods coming from the graph theory field were used to extract patterns in data constituted by links between products and place and date of seizure. A data mining process completed using graphing techniques is called ``graph mining''. Patterns were detected that had to be interpreted and compared with preliminary knowledge to establish their relevancy. The illicit drug profiling process is actually an intelligence process that uses preliminary illicit drug classes to classify new samples. Methods proposed in this study could be used \textit{a priori} to compare structures from preliminary and post-detection patterns. This new knowledge of a repeated structure may provide valuable complementary information to profiling and become a source of intelligence.

Inferring the mode of origin of polyploid species from next-generation sequence data.

Many eukaryote organisms are polyploid. However, despite their importance, evolutionary inference of polyploid origins and modes of inheritance has been limited by a need for analyses of allele segregation at multiple loci using crosses. The increasing availability of sequence data for nonmodel species now allows the application of established approaches for the analysis of genomic data in polyploids. Here, we ask whether approximate Bayesian computation (ABC), applied to realistic traditional and next-generation sequence data, allows correct inference of the evolutionary and demographic history of polyploids. Using simulations, we evaluate the robustness of evolutionary inference by ABC for tetraploid species as a function of the number of individuals and loci sampled, and the presence or absence of an outgroup. We find that ABC adequately retrieves the recent evolutionary history of polyploid species on the basis of both old and new sequencing technologies. The application of ABC to sequence data from diploid and polyploid species of the plant genus Capsella confirms its utility. Our analysis strongly supports an allopolyploid origin of C. bursa-pastoris about 80 000 years ago. This conclusion runs contrary to previous findings based on the same data set but using an alternative approach and is in agreement with recent findings based on whole-genome sequencing. Our results indicate that ABC is a promising and powerful method for revealing the evolution of polyploid species, without the need to attribute alleles to a homeologous chromosome pair. The approach can readily be extended to more complex scenarios involving higher ploidy levels.

Digital evidence, 'absence' of data and ambiguous patterns of reasoning

In this paper we discuss the use of digital data by the Swiss Federal Criminal Court in a recent case of attempted homicide. We use this case to examine drawbacks for the defense when the presentation of scientific evidence is partial, especially when the only perspective mentioned is that of the prosecution. We tackle this discussion at two distinct levels. First, we pursue an essentially non-technical presentation of the topic by drawing parallels between the court's summing up of the case and flawed patterns of reasoning commonly seen in other forensic disciplines, such as DNA and particle traces (e.g., gunshot residues). Then, we propose a formal analysis of the case, using elements of probability and graphical probability models, to justify our main claim that the partial presentation of digital evidence poses a risk to the administration of justice in that it keeps vital information from the defense. We will argue that such practice constitutes a violation of general principles of forensic interpretation as established by forensic science literature and current recommendations by forensic science interest groups (e.g., the European Network of Forensic Science Institutes). Finally, we posit that argument construction and analysis using formal methods can help replace digital evidence appropriately into context and thus support a sound evaluation of the evidence.

Contribution to the geology of Thailand and implications for the geodynamic evolution of Southeast Asia

Abstract The purpose of this study is to unravel the geodynamic evolution of Thailand and, from that, to extend the interpretation to the rest of Southeast Asia. The methodology was based in a first time on fieldwork in Northern Thailand and Southernmost Myanmar, using a multidisciplinary approach, and then on the compilation and re-interpretation, in a plate tectonics point of view, of existing data about the whole Southeast Asia. The main results concern the Nan-Uttaradit suture, the Chiang Mai Volcanic Belt and the proposition of a new location for the Palaeotethys suture. This led to the establishment of a new plate tectonic model for the geodynamic evolution of Southeast Asia, implying the existence new terranes (Orang Laut and the redefinition of Shan-Thai) and the role of the Palaeopacific Ocean in the tectonic development of the area. The model proposed here considers the Palaeotethys suture as located along the Tertiary Mae Yuam Fault, which represents the divide between the Cimmerian Sibumasu terrane and the Indochina-derived Shan-Thai block. The term Shan-Thai, previously used to define the Cimmerian area (when the Palaeotethys suture was thought to represented by the Nan-Uttaradit suture), was redefined here by keeping its geographical location within the Shan States of Myanmar and Central-Northern Thailand, but attributing it an East Asian Origin. Its detachment from Indochina was the result of the Early Permian opening of the Nan basin. The Nan basin closed during the Middle Triassic, before the deposition of Carnian-Norian molasse. The modalities of the closure of the basin imply a first phase of Middle Permian obduction, followed by final eastwards subduction. The Chiang Mai Volcanic Belt consists of scattered basaltic rocks erupted at least during the Viséan in an extensional continental intraplate setting, on the Shan-Thai part of the Indochina block. The Viséan age was established by the dating of limestone stratigraphically overlying the basalts. In several localities of the East Asian Continent, coeval extensional features occur, possibly implying one or more Early Carboniferous extensional events at a regional scale. These events occurred either due to the presence of a mantle plume or to the roll-back of the Palaeopacific Ocean, subducting beneath Indochina and South China, or both. The Palaeopacific Ocean is responsible, during the Early Permian, for the opening of the Song Ma and Poko back-arcs (Vietnam) with the consequent detachment of the Orang Laut Terranes (Eastern Vietnam, West Sumatra, Kalimantan, Palawan, Taiwan). The Late Triassic/Early Jurassic closure of the Eastern Palaeotethys is considered as having taken place by subduction beneath its southern margin (Gondwana), due to the absence of Late Palaeozoic arc magmatism on its northern (Indochinese) margin and the presence of volcanism on the Cimmerian blocks (Mergui, Lhasa). Résumé Le but de cette étude est d'éclaircir l'évolution géodynamique de la Thaïlande et, à partir de cela, d'étendre l'interprétation au reste de l'Asie du Sud-Est. La méthodologie utilisée est basée dans un premier temps sur du travail de terrain en Thaïlande du nord et dans l'extrême sud du Myanmar, en se basant sur une approche pluridisciplinaire. Dans un deuxième temps, la compilation et la réinterprétation de données préexistantes sur l'Asie du Sud-est la été faite, dans une optique basée sur la tectonique des plaques. Les principaux résultats de ce travail concernent la suture de Nan-Uttaradit, la « Chiang Mai Volcanic Belt» et la proposition d'une nouvelle localité pour la suture de la Paléotethys. Ceci a conduit à l'établissement d'un nouveau modèle pour l'évolution géodynamique de l'Asie du Sud-est, impliquant l'existence de nouveaux terranes (Orang Laut et Shan-Thai redéfini) et le rôle joué par le Paléopacifique dans le développement tectonique de la région. Le modèle présenté ici considère que la suture de la Paléotethys est située le long de la faille Tertiaire de Mae Yuam, qui représente la séparation entre le terrain Cimmérien de Sibumasu et le bloc de Shan-Thai, d'origine Indochinoise. Le terme Shan-Thai, anciennement utilise pour définir le bloc Cimmérien (quand la suture de la Paléotethys était considérée être représentée par la suture de Nan-Uttaradit), a été redéfini ici en maintenant sa localisation géographique dans les états Shan du Myanmar et la Thaïlande nord-centrale, mais en lui attribuant une origine Est Asiatique. Son détachement de l'Indochine est le résultat de l'ouverture du basin de Nan au Permien Inférieur. Le basin de Nan s'est fermé pendant le Trias Moyen, avant le dépôt de molasse Carnienne-Norienne. Les modalités de fermeture du basin invoquent une première phase d'obduction au Permien Moyen, suivie par une subduction finale vers l'est. La "Chiang Mai Volcanic Belt" consiste en des basaltes éparpillés qui ont mis en place au moins pendant le Viséen dans un contexte extensif intraplaque continental sur la partie de l'Indochine correspondant au bloc de Shan-Thai. L'âge Viséen a été établi sur la base de la datation de calcaires qui surmontent stratigraphiquement les basaltes. Dans plusieurs localités du continent Est Asiatique, des preuves d'extension plus ou moins contemporaines ont été retrouvées, ce qui implique l'existence d'une ou plusieurs phases d'extension au Carbonifère Inférieur a une échelle régionale. Ces événements sont attribués soit à la présence d'un plume mantellique, ou au rollback du Paléopacifique, qui subductait sous l'Indochine et la Chine Sud, soit les deux. Pendant le Permien inférieur, le Paléopacifique est responsable pour l'ouverture des basins d'arrière arc de Song Ma et Poko (Vietnam), induisant le détachement des Orang Laut Terranes (Est Vietnam, Ouest Sumatra, Kalimantan, Palawan, Taiwan). La fermeture de la Paléotethys Orientale au Trias Supérieur/Jurassique Inférieur est considérée avoir eu lieu par subduction sous sa marge méridionale (Gondwana), à cause de l'absence de magmatisme d'arc sur sa marge nord (Indochinoise) et de la présence de volcanisme sur les blocs Cimmériens de Lhassa et Sibumasu (Mergui). Résumé large public L'histoire géologique de l'Asie du Sud-est depuis environ 430 millions d'années a été déterminée par les collisions successives de plusieurs continents les uns avec les autres. Il y a environ 430 millions d'années, au Silurien, un grand continent appelé Gondwana, a commencé à se «déchirer» sous l'effet des contraintes tectoniques qui le tiraient. Cette extension a provoqué la rupture du continent et l'ouverture d'un grand océan, appelé Paléotethys, éloignant les deux parties désormais séparées. C'est ainsi que le continent Est Asiatique, composé d'une partie de la Chine actuelle, de la Thaïlande, du Myanmar, de Sumatra, du Vietnam et de Bornéo a été entraîné avec le bord (marge) nord de la Paléotethys, qui s'ouvrait petit à petit. Durant le Carbonifère Supérieur, il y a environ 300 millions d'années, le sud du Gondwana subissait une glaciation, comme en témoigne le dépôt de sédiments glaciaires dans les couches de cet âge. Au même moment le continent Est Asiatique se trouvait à des latitudes tropicales ou équatoriales, ce qui permettait le dépôt de calcaires contenant différents fossiles de foraminifères d'eau chaude et de coraux. Durant le Permien Inférieur, il y a environ 295 millions d'années, la Paléotethys Orientale, qui était un relativement vieil océan avec une croûte froide et lourde, se refermait. La croûte océanique a commencé à s'enfoncer, au sud, sous le Gondwana. C'est ce que l'on appelle la subduction. Ainsi, le Gondwana s'est retrouvé en position de plaque supérieure, par rapport à la Paléotethys qui, elle, était en plaque inférieure. La plaque inférieure en subductant a commencé à reculer. Comme elle ne pouvait pas se désolidariser de la plaque supérieure, en reculant elle l'a tirée. C'est le phénomène du «roll-back ». Cette traction a eu pour effet de déchirer une nouvelle fois le Gondwana, ce qui a résulté en la création d'un nouvel Océan, la Neotethys. Cet Océan en s'ouvrant a déplacé une longue bande continentale que l'on appelle les blocs Cimmériens. La Paléotethys était donc en train de se fermer, la Neotethys de s'ouvrir, et entre deux les blocs Cimmériens se rapprochaient du Continent Est Asiatique. Pendant ce temps, le continent Est Asiatique était aussi soumis à des tensions tectoniques. L'Océan Paléopacifique, à l'est de celui-ci, était aussi en train de subducter. Cette subduction, par roll-back, a déchiré le continent en détachant une ligne de microcontinents appelés ici « Orang Laut Terranes », séparés du continent par deux océans d'arrière arc : Song Ma et Poko. Ceux-ci sont composés de Taiwan, Palawan, Bornéo ouest, Vietnam oriental, et la partie occidentale de Sumatra. Un autre Océan s'est ouvert pratiquement au même moment dans le continent Est Asiatique : l'Océan de Nan qui, en s'ouvrant, a détaché un microcontinent appelé Shan-Thai. La fermeture de l'Océan de Nan, il y a environ 230 millions d'années a resolidarisé Shan-Thai et le continent Est Asiatique et la trace de cet événement est aujourd'hui enregistrée dans la suture (la cicatrice de l'Océan) de Nan-Uttaradit. La cause de l'ouverture de l'Océan de Nan peut soit être due à la subduction du Paléopacifique, soit aux fait que la subduction de la Paléotethys tirait le continent Est Asiatique par le phénomène du « slab-pull », soit aux deux. La subduction du Paléopacifique avait déjà crée de l'extension dans le continent Est Asiatique durant le Carbonifère Inférieur (il y a environ 340-350 millions d'années) en créant des bassins et du volcanisme, aujourd'hui enregistré en différents endroits du continent, dont la ceinture volcanique de Chiang Mai, étudiée ici. A la fin du Trias, la Paléotethys se refermait complètement, et le bloc Cimmérien de Sibumasu entrait en collision avec le continent Est Asiatique. Comme c'est souvent le cas avec les grands océans, il n'y a pas de suture proprement dite, avec des fragments de croûte océanique, pour témoigner de cet évènement. Celui-ci est visible grâce à la différence entre les sédiments du Carbonifère Supérieur et du Permieñ Inférieur de chaque domaine : dans le domaine Cimmérien ils sont de type glaciaire alors que dans le continent Est Asiatique ils témoignent d'un climat tropical. Les océans de Song Ma et Poko se sont aussi refermés au Trias, mais eux ont laissé des sutures visibles

D-amphetamine fails to increase extracellular dopamine levels in mice lacking alpha 1b-adrenergic receptors: relationship between functional and nonfunctional dopamine release.

It was found recently that locomotor and rewarding effects of psychostimulants and opiates were dramatically decreased or suppressed in mice lacking alpha1b-adrenergic receptors [alpha1b-adrenergic receptor knock-outs (alpha1bAR-KOs)] (Drouin et al., 2002). Here we show that blunted locomotor responses induced by 3 and 6 mg/kg d-amphetamine in alpha1bAR-KO mice [-84 and -74%, respectively, when compared with wild-type (WT) mice] are correlated with an absence of d-amphetamine-induced increase in extracellular dopamine (DA) levels in the nucleus accumbens of alpha1bAR-KO mice. Moreover, basal extracellular DA levels in the nucleus accumbens are lower in alpha1bAR-KO than in WT littermates (-28%; p < 0.001). In rats however, prazosin, an alpha1-adrenergic antagonist, decreases d-amphetamine-induced locomotor hyperactivity without affecting extracellular DA levels in the nucleus accumbens, a finding related to the presence of an important nonfunctional release of DA (Darracq et al., 1998). We show here that local d-amphetamine releases nonfunctional DA with the same affinity but a more than threefold lower amplitude in C57BL6/J mice than in Sprague Dawley rats. Altogether, this suggests that a trans-synaptic mechanism amplifies functional DA into nonfunctional DA release. Our data confirm the presence of a powerful coupling between noradrenergic and dopaminergic neurons through the stimulation of alpha1b-adrenergic receptors and indicate that nonfunctional DA release is critical in the interpretation of changes in extracellular DA levels. These results suggest that alpha1b-adrenergic receptors may be important therapeutic pharmacological targets not only in addiction but also in psychosis because most neuroleptics possess anti-alpha1-adrenergic properties.

Evaluation of the reconstruction limits of a frequency-independent crosshole georadar waveform inversion scheme in the presence of dispersion

Waveform tomographic imaging of crosshole georadar data is a powerful method to investigate the shallow subsurface because of its ability to provide images of pertinent petrophysical parameters with extremely high spatial resolution. All current crosshole georadar waveform inversion strategies are based on the assumption of frequency-independent electromagnetic constitutive parameters. However, in reality, these parameters are known to be frequency-dependent and complex and thus recorded georadar data may show significant dispersive behavior. In this paper, we evaluate synthetically the reconstruction limits of a recently published crosshole georadar waveform inversion scheme in the presence of varying degrees of dielectric dispersion. Our results indicate that, when combined with a source wavelet estimation procedure that provides a means of partially accounting for the frequency-dependent effects through an "effective" wavelet, the inversion algorithm performs remarkably well in weakly to moderately dispersive environments and has the ability to provide adequate tomographic reconstructions.

Activation of HTLV-I transcription in the presence of Tax is independent of the acetylation of CREB-2 (ATF-4).

We have previously demonstrated that the bZIP transcription factor CREB-2, also called ATF-4, trans-activates, in association with the viral protein Tax, the human T-cell leukemia virus type I (HTLV-I) promoter. In this study, we have examined whether CREB-2 acetylation affects transcriptional activation mediated by Tax. We present evidence that CREB-2 is acetylated in vitro and in vivo. CREB-2 is acetylated in two regions: the basic domain of the bZIP (from amino acid residue 270 to 300) and the short basic domain (from 342 to 351) located downstream from the bZIP. We also demonstrate that CREB-2 is acetylated by p300/CBP but not by p/CAF. Moreover, replacement of lysine by arginine in the basic domains decreases the trans-activating capacity of CREB-2. However, in the presence of Tax, the HTLV-I transcription remains fully activated by these CREB-2 mutants. Although we cannot totally exclude that the mutations could also affect CREB-2 structure and activity independent of acetylation, our results suggest that activation of the viral promoter in the presence of Tax is independent of the CREB-2 acetylation.

A Modulated Closed Form solution for Quantitative Susceptibility Mapping - A thorough evaluation and comparison to iterative methods based on edge prior knowledge.

The aim of this study is to perform a thorough comparison of quantitative susceptibility mapping (QSM) techniques and their dependence on the assumptions made. The compared methodologies were: two iterative single orientation methodologies minimizing the l2, l1TV norm of the prior knowledge of the edges of the object, one over-determined multiple orientation method (COSMOS) and anewly proposed modulated closed-form solution (MCF). The performance of these methods was compared using a numerical phantom and in-vivo high resolution (0.65mm isotropic) brain data acquired at 7T using a new coil combination method. For all QSM methods, the relevant regularization and prior-knowledge parameters were systematically changed in order to evaluate the optimal reconstruction in the presence and absence of a ground truth. Additionally, the QSM contrast was compared to conventional gradient recalled echo (GRE) magnitude and R2* maps obtained from the same dataset. The QSM reconstruction results of the single orientation methods show comparable performance. The MCF method has the highest correlation (corrMCF=0.95, r(2)MCF =0.97) with the state of the art method (COSMOS) with additional advantage of extreme fast computation time. The l-curve method gave the visually most satisfactory balance between reduction of streaking artifacts and over-regularization with the latter being overemphasized when the using the COSMOS susceptibility maps as ground-truth. R2* and susceptibility maps, when calculated from the same datasets, although based on distinct features of the data, have a comparable ability to distinguish deep gray matter structures.

Alpha beta lineage-committed thymocytes can be rescued by the gamma delta T cell receptor (TCR) in the absence of TCR beta chain.

Commitment of the alpha beta and gamma delta T cell lineages within the thymus has been studied in T cell receptor (TCR)-transgenic and TCR mutant murine strains. TCR gamma delta-transgenic or TCR beta knockout mice, both of which are unable to generate TCR alpha beta-positive T cells, develop phenotypically alpha beta-like thymocytes in significant proportions. We provide evidence that in the absence of functional TCR beta protein, the gamma delta TCR can promote the development of alpha beta-like thymocytes, which, however, do not expand significantly and do not mature into gamma delta T cells. These results show that commitment to the alpha beta lineage can be determined independently of the isotype of the TCR, and suggest that alpha beta versus gamma delta T cell lineage commitment is principally regulated by mechanisms distinct from TCR-mediated selection. To accommodate our data and those reported previously on the effect of TCR gamma and delta gene rearrangements on alpha beta T cell development, we propose a model in which lineage commitment occurs independently of TCR gene rearrangement.

Energetics of RecA-mediated recombination reactions. Without ATP hydrolysis RecA can mediate polar strand exchange but is unable to recycle.

We demonstrate that the step of DNA strand exchange during RecA-mediated recombination reaction can occur equally efficiently in the presence or absence of ATP hydrolysis. The polarity of strand exchange is the same when instead of ATP its non-hydrolyzable analog adenosine-5'-O-(3-thiotriphosphate) is used. We show that the ATP dependence of recombination reaction is limited to the post-exchange stages of the reactions. The low DNA affinity state of RecA protomers, induced after ATP hydrolysis, is necessary for the dissociation of RecA-DNA complexes at the end of the reaction. This dissociation of RecA from DNA is necessary for the release of recombinant DNA molecules from the complexes formed with RecA and for the recycling of RecA protomers for another round of the recombination reaction.

Normal hemopoiesis and lymphopoiesis in the combined absence of numb and numblike.

The mammalian ortholog of the conserved Drosophila adaptor protein Numb (Nb) and its homolog Numblike (Nbl) modulate neuronal cell fate determination at least in part by antagonizing Notch signaling. Because the Notch pathway has been implicated in regulating hemopoietic stem cell self-renewal and T cell fate specification in mammals, we investigated the role of Nb and Nbl in hemopoiesis using conditional gene targeting. Surprisingly simultaneous deletion of both Nb and Nbl in murine bone marrow precursors did not affect the ability of stem cells to self-renew or to give rise to differentiated myeloid or lymphoid progeny, even under competitive conditions in mixed chimeras. Furthermore, T cell fate specification and intrathymic T cell development were unaffected in the combined absence of Nb and Nbl. Collectively our data indicate that the Nb family of adaptor proteins is dispensable for hemopoiesis and lymphopoiesis in mice, despite their proposed role in neuronal stem cell development.

Chemical profiling of different hashish seizures by gas chromatography-mass spectrometry and statistical methodology: a case report

Limited information is available regarding the methodology required to characterize hashish seizures for assessing the presence or the absence of a chemical link between two seizures. This casework report presents the methodology applied for assessing that two different police seizures were coming from the same block before this latter one was split. The chemical signature was extracted using GC-MS analysis and the implemented methodology consists in a study of intra- and inter-variability distributions based on the measurement of the chemical profiles similarity using a number of hashish seizures and the calculation of the Pearson correlation coefficient. Different statistical scenarios (i.e., a combination of data pretreatment techniques and selection of target compounds) were tested to find the most discriminating one. Seven compounds showing high discrimination capabilities were selected on which a specific statistical data pretreatment was applied. Based on the results, the statistical model built for comparing the hashish seizures leads to low error rates. Therefore, the implemented methodology is suitable for the chemical profiling of hashish seizures.

Segmentation of brain structures in presence of a space-occupying lesion.

Brain deformations induced by space-occupying lesions may result in unpredictable position and shape of functionally important brain structures. The aim of this study is to propose a method for segmentation of brain structures by deformation of a segmented brain atlas in presence of a space-occupying lesion. Our approach is based on an a priori model of lesion growth (MLG) that assumes radial expansion from a seeding point and involves three steps: first, an affine registration bringing the atlas and the patient into global correspondence; then, the seeding of a synthetic tumor into the brain atlas providing a template for the lesion; finally, the deformation of the seeded atlas, combining a method derived from optical flow principles and a model of lesion growth. The method was applied on two meningiomas inducing a pure displacement of the underlying brain structures, and segmentation accuracy of ventricles and basal ganglia was assessed. Results show that the segmented structures were consistent with the patient's anatomy and that the deformation accuracy of surrounding brain structures was highly dependent on the accurate placement of the tumor seeding point. Further improvements of the method will optimize the segmentation accuracy. Visualization of brain structures provides useful information for therapeutic consideration of space-occupying lesions, including surgical, radiosurgical, and radiotherapeutic planning, in order to increase treatment efficiency and prevent neurological damage.

Neuropathological substrates and structural changes in late-life depression: the impact of vascular burden.

A first episode of depression after 65 years of age has long been associated with both severe macrovascular and small microvascular pathology. Among the three more frequent forms of depression in old age, post-stroke depression has been associated with an abrupt damage of cortical circuits involved in monoamine production and mood regulation. Late-onset depression (LOD) in the absence of stroke has been related to lacunes and white matter lesions that invade both the neocortex and subcortical nuclei. Recurrent late-life depression is thought to induce neuronal loss in the hippocampal formation and white matter lesions that affect limbic pathways. Despite an impressive number of magnetic resonance imaging (MRI) studies in this field, the presence of a causal relationship between structural changes in the human brain and LOD is still controversial. The present article provides a critical overview of the contribution of neuropathology in post-stroke, late-onset, and late-life recurrent depression. Recent autopsy findings challenge the role of stroke location in the occurrence of post-stroke depression by pointing to the deleterious effect of subcortical lacunes. Despite the lines of evidences supporting the association between MRI-assessed white matter changes and mood dysregulation, lacunes, periventricular and deep white matter demyelination are all unrelated to the occurrence of LOD. In the same line, neuropathological data show that early-onset depression is not associated with an acceleration of aging-related neurodegenerative changes in the human brain. However, they also provide data in favor of the neurotoxic theory of depression by showing that neuronal loss occurs in the hippocampus of chronically depressed patients. These three paradigms are discussed in the light of the complex relationships between psychosocial determinants and biological vulnerability in affective disorders.