Decision-theoretic planning is a popular approach to sequential decision making problems, because it treats uncertainty in sensing and acting in a principled way. In single-agent frameworks like MDPs and POMDPs, planning can be carried out by resorting to Q-value functions: an optimal Q-value function Q* is computed in a recursive manner by dynamic programming, and then an optimal policy is extracted from Q*. In this paper we study whether similar Q-value functions can be defined for decentralized POMDP models (Dec-POMDPs), and how policies can be extracted from such value functions. We define two forms of the optimal Q-value function for Dec-POMDPs: one that gives a normative description as the Q-value function of an optimal pure joint policy and another one that is sequentially rational and thus gives a recipe for computation. This computation, however, is infeasible for all but the smallest problems. Therefore, we analyze various approximate Q-value functions that allow for efficient computation. We describe how they relate, and we prove that they all provide an upper bound to the optimal Q-value function Q*. Finally, unifying some previous approaches for solving Dec-POMDPs, we describe a family of algorithms for extracting policies from such Q-value functions, and perform an experimental evaluation on existing test problems, including a new firefighting benchmark problem.
Abstract Background In 2012, Luxembourg introduced a Referring Doctor (RD) policy, whereby patients voluntarily register with a primary care practitioner, who coordinates patients’ health care and ensures optimal follow-up. We contribute to the limited evidence base on patient registration by evaluating the effects of the RD policy. Methods We used data on 16,775 people with type 2 diabetes on oral medication (PWT2D), enrolled with the Luxembourg National Fund from 2010 to 2018. We examined the utilisation of primary and specialist outpatient care, quality of care process indicators, and reimbursed prescribed medicines over the short- (until 2015) and medium-term (until 2018). We used propensity score matching to identify comparable groups of patients with and without an RD. We applied difference-in-differences methods that accounted for patients’ registration with an RD in different years. Results There was low enrolment of PWT2D in the RD programme. The differences-in-differences parallel trends assumption was not met for: general practitioner (GP) consultations, GP home visits (medium-term), HbA1c test (short-term), complete cholesterol test (short-term), kidney function (urine) test (short-term), and the number of repeat prescribed cardiovascular system medicines (short-term). There was a statistically significant increase in the number of: HbA1c tests (medium-term: 0.09 (95% CI: 0.01 to 0.18)); kidney function (blood) tests in the short- (0.10 (95% CI: 0.01 to 0.19)) and medium-term (0.11 (95% CI: 0.03 to 0.20)); kidney function (urine) tests (medium-term: 0.06 (95% CI: 0.02 to 0.10)); repeat prescribed medicines in the short- (0.19 (95% CI: 0.03 to 0.36)) and medium-term (0.18 (95% CI: 0.02 to 0.34)); and repeat prescribed cardiovascular system medicines (medium-term: 0.08 (95% CI: 0.01 to 0.15)). Sensitivity analyses also revealed increases in kidney function (urine) tests (short-term: 0.07 (95% CI: 0.03 to 0.11)) and dental consultations (short-term: 0.06, 95% CI: 0.00 to 0.11), and decreases in specialist consultations (short-term: -0.28, 95% CI: -0.51 to -0.04; medium-term: -0.26, 95% CI: -0.49 to -0.03). Conclusions The RD programme had a limited effect on care quality indicators and reimbursed prescribed medicines for PWT2D. Future research should extend the analysis beyond this cohort and explore data linkage to include clinical outcomes and socio-economic characteristics.
The protein dynamics are common and important for their biological functions and properties, the study of which usually involves time-consuming molecular dynamics (MD) simulations in silico. Recently, generative models has been leveraged as a surrogate sampler to obtain conformation ensembles with orders of magnitude faster and without requiring any simulation data (a "zero-shot" inference). However, being agnostic of the underlying energy landscape, the accuracy of such generative model may still be limited. In this work, we explore the few-shot setting of such pre-trained generative sampler which incorporates MD simulations in a tractable manner. Specifically, given a target protein of interest, we first acquire some seeding conformations from the pre-trained sampler followed by a number of physical simulations in parallel starting from these seeding samples. Then we fine-tuned the generative model using the simulation trajectories above to become a target-specific sampler. Experimental results demonstrated the superior performance of such few-shot conformation sampler at a tractable computational cost.
A bstractWe introduce a new framework for constructing black hole solutions that are holographically dual to strongly coupled field theories with explicitly broken translation invariance. Using a classical gravitational theory with a continuous global symmetry leads to constructions that involve solving ODEs instead of PDEs. We study in detail D = 4 Einstein-Maxwell theory coupled to a complex scalar field with a simple mass term. We construct black holes dual to metallic phases which exhibit a Drude-type peak in the optical conductivity, but there is no evidence of an intermediate scaling that has been reported in other holographic lattice constructions. We also construct black holes dual to insulating phases which exhibit a suppression of spectral weight at low frequencies. We show that the model also admits a novel AdS3 × $ \mathbb{R} $ solution.
Neutron Star Measurements Neutron stars are one of the densest manifestations of matter in the universe. Yagi and Yunes (p. 365) examined the moment of inertia of neutron stars, which determines how fast they can spin, and the quadrupole moment and tidal Love number, which determine how much they can be deformed. The findings suggest that these three quantities obey universal relationships that are independent of the internal structure of the stars, implying that measurements of one of the three could accurately predict the other two. The relation of inertia, Love number, and quadrupole moment is independent of neutron and quark stars’ internal structure. Neutron stars and quark stars are not only characterized by their mass and radius but also by how fast they spin, through their moment of inertia, and how much they can be deformed, through their Love number and quadrupole moment. These depend sensitively on the star’s internal structure and thus on unknown nuclear physics. We find universal relations between the moment of inertia, the Love number, and the quadrupole moment that are independent of the neutron and quark star’s internal structure. These can be used to learn about neutron star deformability through observations of the moment of inertia, break degeneracies in gravitational wave detection to measure spin in binary inspirals, distinguish neutron stars from quark stars, and test general relativity in a nuclear structure–independent fashion.
Protein-DNA interaction is critical for life activities such as replication, transcription, and splicing. Identifying protein-DNA binding residues is essential for modeling their interaction and downstream studies. However, developing accurate and efficient computational methods for this task remains challenging. Improvements in this area have the potential to drive novel applications in biotechnology and drug design. In this study, we propose a novel approach called CLAPE, which combines a pre-trained protein language model and the contrastive learning method to predict DNA binding residues. We trained the CLAPE-DB model on the protein-DNA binding sites dataset and evaluated the model performance and generalization ability through various experiments. The results showed that the AUC values of the CLAPE-DB model on the two benchmark datasets reached 0.871 and 0.881, respectively, indicating superior performance compared to other existing models. CLAPE-DB showed better generalization ability and was specific to DNA-binding sites. In addition, we trained CLAPE on different protein-ligand binding sites datasets, demonstrating that CLAPE is a general framework for binding sites prediction. To facilitate the scientific community, the benchmark datasets and codes are freely available at https://github.com/YAndrewL/clape.
Aryo Pradipta Gema, Michał Kobiela, Achille Fraisse
et al.
The SARS-CoV-2 pandemic has emphasised the importance of developing a universal vaccine that can protect against current and future variants of the virus. The present study proposes a novel conditional protein Language Model architecture, called Vaxformer, which is designed to produce natural-looking antigenicity-controlled SARS-CoV-2 spike proteins. We evaluate the generated protein sequences of the Vaxformer model using DDGun protein stability measure, netMHCpan antigenicity score, and a structure fidelity score with AlphaFold to gauge its viability for vaccine development. Our results show that Vaxformer outperforms the existing state-of-the-art Conditional Variational Autoencoder model to generate antigenicity-controlled SARS-CoV-2 spike proteins. These findings suggest promising opportunities for conditional Transformer models to expand our understanding of vaccine design and their role in mitigating global health challenges. The code used in this study is available at https://github.com/aryopg/vaxformer .
Deep generative models have recently achieved superior performance in 3D molecule generation. Most of them first generate atoms and then add chemical bonds based on the generated atoms in a post-processing manner. However, there might be no corresponding bond solution for the temporally generated atoms as their locations are generated without considering potential bonds. We define this problem as the atom-bond inconsistency problem and claim it is the main reason for current approaches to generating unrealistic 3D molecules. To overcome this problem, we propose a new diffusion model called MolDiff which can generate atoms and bonds simultaneously while still maintaining their consistency by explicitly modeling the dependence between their relationships. We evaluated the generation ability of our proposed model and the quality of the generated molecules using criteria related to both geometry and chemical properties. The empirical studies showed that our model outperforms previous approaches, achieving a three-fold improvement in success rate and generating molecules with significantly better quality.
Many crucial biological processes rely on networks of protein-protein interactions. Predicting the effect of amino acid mutations on protein-protein binding is vital in protein engineering and therapeutic discovery. However, the scarcity of annotated experimental data on binding energy poses a significant challenge for developing computational approaches, particularly deep learning-based methods. In this work, we propose SidechainDiff, a representation learning-based approach that leverages unlabelled experimental protein structures. SidechainDiff utilizes a Riemannian diffusion model to learn the generative process of side-chain conformations and can also give the structural context representations of mutations on the protein-protein interface. Leveraging the learned representations, we achieve state-of-the-art performance in predicting the mutational effects on protein-protein binding. Furthermore, SidechainDiff is the first diffusion-based generative model for side-chains, distinguishing it from prior efforts that have predominantly focused on generating protein backbone structures.