256488 research outputs found
Sort by
The RNA Recognition Motif (RRM) of the Arabidopsis RS2Z32 and RS2Z33 splicing factors coordinates protein-protein and protein-RNA interactions, and contributes to their nucleocytoplasmic dynamics.
No full textpeer reviewedThe Arabidopsis splicing factors arginine/serine-rich zinc knuckle-containing proteins 32 and 33 (RS2Z32 and RS2Z33) are plant-specific members of the SR family. Here, we characterized both RS2Z32 and RS2Z33 by examining their expression profile at different stages of development and their spatial cellular distribution, as well as the contribution of their domains to the establishment of protein-protein interactions and to RNA binding specificity. We report that the RS2Z32 and RS2Z33 promoters are ubiquitously active during vegetative and reproductive growth, and that both RS2Z splicing factors localize in the nucleus (except the nucleolus). We show that the C-terminal arginine/serine-rich (RS) domain, but not the serine/proline-rich (SP) extension, is a determinant of nuclear localization. We demonstrate that their RNA recognition motif (RRM) domain specifically binds pyrimidine-rich RNA motifs via three residues (Y14, Y46, F48), and is also involved in protein-protein interactions with at least three SR proteins, namely SR45, SCL30, and SR34. Finally, we show that mutations in RNA-binding domains (i.e. RRM and zinc knuckles, ZnKs) affect the nucleocytoplasmic dynamics of both RS2Z proteins. Our findings provide molecular evidence for the involvement of plant-specific SR splicing factors into the regulation of the splicing process
De la présence de l'épopée occitane dans quelques ouvrages de littérature médiévale
No full textL'objectif de cette contribution est d'examiner la place réservée à l'épopée occitane et le traitement dont fait l'objet cette matière dans des manuels d'histoire de la littérature médiévale française et/ou occitane, ainsi que dans quelques anthologies.EPOC-S (projet financé APEX 2025
Local product cooperatives in Morocco: between heritage promotion and marketing challenges
No full textpeer reviewe
Multivariate functional isolation forest for money laundering detection
No full textMoney laundering is a complex sequence of money transfers that traditional methods based on aggregated transactions struggle to detect. To improve upon this issue, the detection of such anomalies is framed as identifying unusual (co-) movements across dimensions in time and scale, with interactions between pairs of dimensions quantified using several projection-based methods. Then, this information is embedded in an isolation forest algorithm extended to multivariate functional data by using these interaction scores as splitting criteria, generating an anomaly score. Finally, this anomaly score ranks observations by their deviation from a functional baseline and allows the identification of suspicious transactions. Empirically, this method is applied to a large sample of bank transaction records, where customers' behaviors are represented as multivariate transaction curves across cash, wire, and international credit/debit dimensions, highlighting money laundering cases overlooked by conventional methods. Depth-based global and local feature importance metrics are used to identify the most informative dimensions. Evaluation against functional and tabular baselines confirms the effectiveness of the proposed methodology. In addition, the performance of the methodology is showcased in a series of realistic simulations
Functional Features for Money Laundering Detection in a Retail Bank
No full textMoney laundering is a complex sequence of money transfers that traditional methods based on aggregated transactions struggle to detect. To improve upon this issue, the detection of such anomalies is framed as identifying unusual (co-) movements across dimensions in time and scale, with interactions between pairs of dimensions quantified using several projection-based methods.
Then, these multivariate functional interactions scores are embedded in unsupervised algorithms (isolation forest, KNN and LOF) generating an anomaly score.
Finally, this anomaly score ranks observations by their deviation from a functional baseline and allows the identification of suspicious transactions. Empirically, this methodology is applied to a large sample of bank transaction records, where customers' behaviors are represented as multivariate transaction curves across cash, wire, and international credit/debit dimensions, highlighting money laundering cases overlooked by conventional methods.
Evaluation against functional and tabular baselines confirms the effectiveness of the proposed methodology. In addition, the performance of the methodology is showcased in a series of realistic simulations
Towards a nonsmooth multibody finite element framework for braiding simulation
No full textHighly deformable slender structures such as cables and hoses are ubiquitous in engineering systems, where they interact with each other and rigid bodies through kinematic joints and contact conditions. The global dynamics of such assemblies can be captured using system-level simulations. However this is not straightforward, as rigid bodies demand a robust handling of large amplitude motions and constrained relative motions, while for flexible bodies, these overall motions are accompanied by local deformations. In addition, nonsmooth phenomena can be induced from impacts or stick-slip transitions. Such factors make the model susceptible to numerical ill-conditioning and stability issues; all of which pose as central problems in the targeted application of this thesis. This thesis proposes two original contributions. (i.) First, we address the simulation of the braiding process, which is widely used in the fabrication of preforms for composite production. On the one end, thin and flexible textile yarns flow from bobbin carriers in a controlled manner, driven by horn gears to form an interlocked mesh. On the other end, yarns undergo contact-friction interactions with the mandrel. As the process parameters strongly influence the final preform geometry, a system-level model is useful. We propose a multibody approach, where the carriers and the mandrel are modelled as rigid bodies, and the yarns are modelled using a geometrically exact beam finite element model. Absolute coordinates represent both large rigid body motions and elastic deflections of the yarns. A local frame representation of the equations of motion leads to reduced geometric nonlinearities. Phenomena such as tension compensation in the yarn are modelled using judicious choices of kinematic joints, and carrier motions using switching bilateral constraints. Furthermore, unilateral contact constraints and friction laws govern the yarn-to-mandrel interactions. We explore two possible contact formulations: a mortar line-to-line contact formulation, and a collocation node-to-surface method. A strict enforcement of constraints using Lagrange multipliers induces a nonsmooth response. Numerical discretization in space is performed using a Lie group interpolation scheme. To solve the semi-discrete system, we use the nonsmooth generalized- method (NSGA) method. The choice of the time step solver in the NSGA is also addressed. An augmented Lagrangian formulation with a semi-smooth Newton method relies on the tangent operator, which becomes ill-posed when the density of contact points becomes large. Instead, we use a nonlinear block Gauss-Seidel strategy which considers a sequential resolution of each contact point. Finally, we present the preliminary simulation of an industrial multi-layer braiding machine. (ii.) Second, the novel approach of switching bilateral constraints for carrier motions provided motivation to address a more fundamental question: \textit{can we establish a general modelling procedure for multibody systems with switching bilateral constraints}? In this work, we address a special class of multibody systems, where switching functions trigger instantaneous changes in the algebraic constraint expressions, thus leading to a time-discontinuous response. These switching functions define switching surfaces, that partition the system dynamics into discrete regimes with nonsmooth transitions. Switching surfaces are thus instrumental in our modelling framework, as they orchestrate the geometry of the constraint space. The equations of motion can either be formulated as hybrid differential-algebraic equations or as an equality of differential measures with constraints. At the switching surface, an impact law based on an intermediate gradient is introduced. We propose to determine this intermediate gradient by interpolation between the pre- and post-switch gradients. Theoretical arguments and numerical results show that the choice of this intermediate gradient drives the energy behavior at switching. The numerical integration demands special attention, as classical DAE solvers fail to handle discontinuities at switching surfaces. Event-driven and time-stepping schemes from nonsmooth dynamics are well-suited for this class of problems. We use a benchmark test to compare the solutions obtained from classical and nonsmooth versions of the generalized- method. Further, using the NSGA scheme, three examples of multibody systems with switching bilateral constraints are successfully simulated.THREA
Alternative crops for climate adaptation in low-lying regions: The case of lotus farming in Thua Thien Hue Province, Central Vietnam
No full textpeer reviewedCrop conversion from rice to alternative crops has gained momentum in many Southeast Asian countries as a strategy to adapt to climate change and strengthen rural livelihoods. This research examines lotus farming in Thua Thien Hue Province, one of the nation’s most climate-vulnerable regions. We used a mixed-methods approach that included key informant interviews, focus group discussions, and a household survey with 95 lotus farmers. The shift to lotus cultivation was driven by the poor performance of rice in low-lying areas and farmers’ expectations of higher returns. Crop conversion policies also improved land access for many households. Although lotus is considered a high-value crop and plays an important role in land-use strategies for flood-prone areas, benefit–cost analysis showed it was economically unviable for most households in 2024. Smallholders relying on family labor tended to earn higher returns, creating a paradox where larger-scale operations did not improve profitability. Repeated crop failures from nematode infestations and extreme weather, combined with high production costs, created significant constraints, particularly for larger farms. Improving the sustainability of lotus and other alternative crops will require promoting indigenous lotus varieties, expanding mechanization, and strengthening post-harvest practices to increase efficiency and resilience
Farmers’ Perception of Ecosystem Services Provided by Historical Rubber Plantations in Sankuru Province, DR Congo
No full textpeer reviewedThe province of Sankuru, located within the Democratic Republic of Congo, is distinguished by its extensive rubber plantations, which have a long history in the region. These plantations have had a considerable impact on the region’s agrarian landscape over time. In addition to the exploitation of latex, for which the conditions are currently very limited, these plantations provide goods and services to the local population and are dominated by rural communities that are highly dependent on these natural resources. This study aimed to characterize the socio-demographic and agrarian profile of historical rubber plantations while assessing the occurrence of the ecosystem services (ESs) they provide. Particular attention will be paid to the farmers’ perceptions of these services, an essential element for the rational management of natural resources. This study used a mixed methodological approach, integrating semi-structured interviews, focus groups, and statistical analyses including chi-square testing and multiple correspondence factorial analysis (MCAFA) to obtain and analyze the data comprehensively. The results indicate that historical rubber plantations in Sankuru provide 21 ESs, which are grouped into four categories: eleven provisioning services, four regulating services, four cultural services, and two supporting services. It has been observed that local communities attach significant importance to the provision of services including the provision of firewood (96.67%) and the utilization of forest resources for traditional pharmacopoeia (91.33%). These plantations have come to be regarded as valuable cultural heritage by local communities over time. The younger generation evinces a greater interest in utility services than the older generation, which displays a preference for cultural services. However, older people demonstrate a more profound understanding of cultural and regulatory services. By emphasizing the species that contribute to ESs and recognizing plantations as cultural heritage, the study enhances the comprehension of the significance of local ecosystems. These findings provide a crucial foundation for directing local policy toward integrated management of historic rubber plantations in Sankuru. By considering the perceptions of local people, the study contributes to the sustainable conservation of these plantations for the present and future generations
Community Perceptions and Determinants of the Sustained Conservation of Historical Rubber Plantations in the Lomela and Lodja Territories, Sankuru Province, Democratic Republic of the Congo
No full textpeer reviewedThe drastic and historic fall in natural rubber prices has prompted many smallholders around the world to abandon rubber plantations in favor of other survival alternatives. In the Lomela and Lodja territories of Sankuru Province (Democratic Republic of Congo), where a historical absence of a rubber market prevails, local communities have conserved rubber plantations inherited from the colonial era (dating back to 1955). Data collected from 401 households enabled us to identify the perceptions and determinants of rubber plantation conservation. The results show that households are highly dependent on forest ecosystem services. Agriculture is the main activity for 81.3% of respondents, in the context of extreme poverty where daily incomes amount to 0.33 USD/person. The patriarchal system favored men, who inherited 97% of the plantations. Men perceived the conservation of the plantations as beneficial, while women perceived it as serving external project interests. Perceptions were significantly influenced by gender, age, social and legal organization, geographical origin, mode of acquisition, main activity, diversification of income sources, membership in a tribal mutuality, access to the informal mutual aid networks, membership in an association and contact with extension services. Conservation was positively and significantly correlated with geographical origin, membership in an association, contact with extension service, consideration of plantations as natural heritage and the ecosystem services provided. These results underline that rubber plantations cannot be understood only in terms of rubber production, but also in terms of their socio-ecological and heritage dimensions