This research introduces a web-based digital twin model designed to predict drug resistance associated with CYP3A7 variants in the fetal liver. As a key enzyme in fetal hepatic metabolism, CYP3A7 significantly influences the pharma-cokinetics of various drugs during development. Genetic variants of CYP3A7 can alter its activity, impacting drug safety and efficacy. To support personalized treatment strategies, the model integrates MichaelisMenten kinetics and physiologically relevant fetal parameters such as enzyme expression levels, liver volume, and gestational age to simulate drug concentration dynamics over time. The platform compares wild-type and resistant enzyme behaviors to evaluate toxicity risk. Our results show that drugs like midazolam remain above toxicity thresholds significantly longer in the CYP3A7*3 variant compared to CYP3A7*2, em-phasizing the critical impact of enzyme polymorphisms on fetal drug exposure. These findings highlight the potential of digital twin models to inform safer pharmacological decisions during pregnancy.

Proteins are the fundamental units of all living cells and play a vital role in various cellular functions. Moreover, proteins are the building blocks of all diseases, making them imperative to the drug discovery process. The ability to identify proteins' characteristics and properties enables scientists with in-depth knowledge of the mechanisms of a disease, that can alter protein behavior. One important aspect in studying proteins is their structure flexibility since the changes of amino acid residue could change the size, shape, and its property. Moreover, proteins usually have a large number of residues and atoms, the approaches considering protein structures are computationally expensive. In this work, we address these problems by combining machine learning approaches that are able to process high dimensional data and feature analyzing techniques to extract underlying relationships within the data. Specifically, we propose a novel metric to capture the locality of the protein structures and a modified transformer architecture to take advantage of both 3D and string representations of proteins. The results show that our model has a comparable performance with state-of-the-art models on the selected tasks from MoleculeNet.

Failure of any component in a robotic system during operation is a critical concern, and it is essential to address such incidents promptly. This work investigates a novel technique to recover from failures or changes in the configuration space while avoiding expensive re-computation or re-planning. We propose the Minimal Path Violation (MPV) concept to find the best feasible path with minimal re-configurations. The algorithm ranks pathways based on visibility, expansiveness, and cost. We perform experiments with articulated 3 DOF to 28 DOF robots ranging from serial linkage robots, Kuka YouBots, and PR2 robots. Our results show that our method outperforms existing optimal planners in computation time, total nodes, and path cost while preserving path feasibility in changed configuration space.

Malaria is a mosquito-borne disease caused by single-celled blood parasites of the genus Plasmodium. In this work, we proposed semi-supervised machine learning models for predicting anti-malaria drugs. Our model consists of two-stage procedures: pre-training and finetuning. Our model takes advantage of a large amount of unlabeled and label data for molecular properties prediction. During the pre-training stage, we incorporate the Masked Language Model, which is widely used in natural language processing for learning text representations, to learn a representational topology of molecular chemical space. Then in the fine-tuning stage, our model is trained on a smaller dataset with the label to perform downstream tasks such as classification or regression. Initial results show that our model has a comparable performance with state-of-the-art models on the selected downstream tasks from MoleculeNet.

Intrinsically disordered proteins (IDPs) play vital regulatory roles in biology, emphasizing the significance of understanding their conformational behavior and interaction mechanisms during protein-ligand or protein-protein interactions. However, IDP analysis becomes difficult due to the lack of a stable structure. In this work, we investigate the binding behavior of an IDP using the surface information of the interacting protein complex. Our algorithm extracts the protein surface model’s topological and geometric features and predicts a geometrically favorable binding pose for an IDP around it. A transition path is planned to the predicted bound position to help evaluate the RMSD deviation in the IDP conformation structure. In our results, we use the Zernike descriptor metric to examine the structural homology of the binding pose and analyze the molar Gibbs free energy (binding affinity) of experimental conformation.

We present an incremental topology-based motion planner that, while planning paths in the configuration space, performs metric gluing on the constructed Vietoris-Rips simplicial complex of each sub-space (voxel). By incrementally capturing topological and geometric information in batches of voxel graphs, our algorithm avoids the time overhead of analyzing the properties of the entire configuration space. We theoretically prove in this paper that the simplices of all voxel graphs joined together are homotopy-equivalent to the union of the simplices in the configuration space. Experiments were carried out in seven different environments using various robots, including the articulated linkage robot, the Kuka YouBot, and the PR2 robot. In all environments, the results show that our algorithm achieves better convergence for path cost and computation time with a memory-efficient roadmap than state-of-the-art methods.

Geometric features of protein surfaces play an important role in the identification of biomolecular structures, functions, and interactions. These features have been crucial in predicting binding sites for protein-ligand or protein-protein interactions. This paper introduces simplicial complexes and discrete Morse theory to extract important geometric information on the protein surface. Using the extracted geometric information, we provide possible intermediate conformations around the protein surface as the ligand travels to the binding site. We compare the efficiency of our method with the state-of-art-method in terms of computation time and total complexes needed to generate the topological structure of the protein surface. We also show comparable relevance of the binding affinity of our method in relation to the known native protein binding site.

Many state-of-art robotics applications require fast and efficient motion planning algorithms. Existing motion planning methods become less effective as the dimensionality of the robot and its workspace increases, especially the computational cost of collision detection routines. In this work, we present a framework to address the cost of expensive primitive operations in sampling-based motion planning. This framework determines the validity of a sample robot configuration through a novel combination of a Contractive AutoEncoder (CAE), which captures an occupancy grids representation of the robot's workspace, and a Multilayer Perceptron (MLP), which efficiently predicts the collision state of the robot using the output from the CAE. We evaluate our framework on multiple planning problems with a variety of robots in 2D and 3D workspaces. The results show that (1) the framework is computationally efficient in all investigated problems, and (2) the framework generalizes well to new workspaces.

We present a topological method for finding coarsely diverse pathways. The use of pre-computed paths for online planning in a dynamic context reduces the overhead of re-planning alternate routes. Our algorithm applied the notion of discrete Morse theory to identify critical points incident on the obstacles and used this information to identify and return a diverse set of coarse paths. Three sampling-based planning approaches are converted to topology-aware planners and compared to another that employs the SPARS2 path planning algorithm. We report on the number of coarse pathways found, computation time, and average path length and show that our approach outperformed previously published path diversity algorithms.

Malaria is a disease caused by single-celled blood parasites of the genus Plasmodium. The protozoan Plasmodium Falciparum (PF) inflicts the most damage and is responsible for most malaria-related deaths. The high mutational capacity of the Plasmodium parasite coupled with its changing metabolism makes the development of new effective drug treatments an evolving and open problem. In this work, we propose a machine learning approach to predict the binding pose structure of ligand families for this parasite. Identifying appropriate protein-ligand binding poses is essential in structure-based drug design and important for the evaluation of protein-ligand binding affinity. Specifically, a conditional variational autoencoder is trained to learn the distribution which represents the binding structures conditioned on the given binding sites. Using this well-trained conditional variational autoencoder, our approach generates binding poses for the ligand's receptor for a particular binding site by sampling from this learned distribution. We demonstrate that our model is able to accurately predict the binding structures of multiple binding sites for the PF parasite invasion ligand families in the erythrocyte invasion and compare it with other states of the art methods like HADDOCK, Hdock, and GRAMMX.

Successful robot path planning is challenging in the presence of visual occlusions and moving targets. Classical methods to solve this problem have used visioning and perception algorithms in addition to partially observable markov decision processes to aid in path planning for pursuit-evasion and robot tracking. We present a predictive path planning process that measures and utilizes the uncertainty present during robot motion planning. We develop a variant of subjective logic in combination with the Markov decision process (MDP) and provide a measure for belief, disbelief, and uncertainty in relation to feasible trajectories being generated. We then model the MDP to identify the best path planning method from a list of possible choices. Our results show a high percentage accuracy based on the closest acquired proximity between a target and a tracking robot and a simplified pursuer trajectory in comparison with related work.

Malaria is a mosquito-borne disease caused by single-celled blood parasites of the genus Plasmodium. The most severe cases of this disease are caused by the Plasmodium species, Falciparum. Once infected, a human host experiences symptoms of recurrent and intermittent fevers occurring over a timeframe of 48 hours, attributed to the synchronized developmental cycle of the parasite during the blood stage. To understand the regulated periodicity of Plasmodium Falciparum (P.F) transcription, this paper implements a well-tuned recurrent neural network with gated recurrent units to forecast and predict the Plasmodium Falciparum gene transcription during its blood stage life cycle. We provide results of this prediction on 697 genes and in particular those genes that express possible drug target enzymes. Our results show a high level of accuracy in being able to predict and forecast the expression levels of the different genes.

This paper introduces a probabilistic prediction model with a novel variant of the Markov decision process to improve tracking time and location detection accuracy in an air-to-ground robot surveillance scenario. While most surveillance algorithms focus mainly on controls of an unmanned aerial vehicle (UAV) and camera for faster tracking of an unmanned ground vehicle (UGV), this paper proposes a way of minimizing detection and tracking time by applying a prediction model to the first observed path taken by the UGV. We present a pursuit algorithm that addresses the problem of target (UGV) localization by combining prediction of used planning algorithm by the target, and application of the same planning algorithm to predict future trajectories. Our results show a high predictive accuracy based on a final position attained by the target and the location predicted by our model.

We present a new way of constructing sparse roadmaps using point clouds that approximates and measures the underlying topology of the Cfreespace. The main advantage of the constructed roadmap is its homotopy equivalence to the η-offset of the Cfreespace. Though only used to plan paths as a regular roadmap in this work, because the roadmap preserves the topology of the underlying sampled space, the information can be used to plan paths beyond the simple connection of graph vertices. To construct the roadmap, we first sample the configuration space so that the resulting graph is a n-skeleton graph that constructs a Vietoris-Rips (VR) complex. Then, we perform a series of topological collapses to remove vertices from the graph while still preserving its topological properties. The resulting roadmaps are used to plan paths for different robots and the experimental results show that the proposed topological approach is faster and more feasible in complex high-dimensional spaces.

Malaria is one of the world's serious diseases causing death of about half a million people in 2015. The protozoan Plasmodium Falciparum inflicts the most damage and is responsible for most malaria related deaths. Biomedical research could enable treating the disease by effectively and specifically targeting essential enzymes of this parasite. However, the parasite has developed resistance to existing drugs, thus making it essential to discover new drugs. We have established a simple computational tool which analyses the topology of the metabolic network of Plasmodium Falciparum to identify essential enzymes as possible drug targets. We investigated the importance of an enzyme in the metabolic network by deleting (knocking-out) a reaction in simulation and examining its effect on the remaining network. Our algorithm then checked whether neighboring compounds of the investigated reaction could be produced by alternative biochemical pathways by using breadth first searches on the whole network. We proposed the use of evolutionary distances as a feature for our machine learning approach to identify potential drug targets. With the help of the machine learning method with the extracted features, we validated previously confirmed and published drug targets in metabolic network of Plasmodium Falciparum. We further identified two new potential targets: dihydrolipoyl dehydrogenase and aconitate hydratase using our approach.

Motivation: Simulating protein folding motions is an important problem in computational biology. Motion planning algorithms such as Probabilistic Roadmap Methods (PRMs) have been successful in modeling the protein folding landscape. PRMs and variants contain several phases (i.e., sampling, connection, and path extraction). Global machine learning has been applied to the connection phase but is inefficient in situations with varying topology, such as those typical of folding landscapes. Results: We present a local learning algorithm that considers the past performance near the current connection attempt as a basis for learning. It is sensitive not only to different types of landscapes but also to differing regions in the landscape itself, removing the need to explicitly partition the landscape. We perform experiments on 23 proteins of varying secondary structure makeup with 52-114 residues. Our method models the landscape with better quality and comparable time to the best performing individual method and to global learning.

Probabilistic Roadmap Methods (PRMs) solve the motion planing problem by constructing a roadmap (or graph) that models the motion space when feasible local motions exist. PRMs and variants contain several phases during roadmap generation i.e., sampling, connection, and query. Some work has been done to apply machine learning to the connection phase to decide which variant to employ, but it uses a global learning approach that is inefficient in heterogeneous situations. We present an algorithm that instead uses local learning: it only considers the performance history in the vicinity of the current connection attempt and uses this information to select good candidates for connection. It thus removes any need to explicitly partition the environment which is burdensome and typically difficult to do. Our results show that our method learns and adapts in heterogeneous environments, including a KUKA youBot with a fixed and mobile base. It finds solution paths faster for single and multi-query scenarios and builds roadmaps with better coverage and connectivity given a fixed amount of time in a wide variety of input problems. In all cases, our method outperforms the previous adaptive connection method and is comparable or better than the best individual method.

Probabilistic Roadmap Methods (PRMs) are widely used motion planning methods that sample robot configurations (nodes) and connect them to form a graph (roadmap) containing feasible trajectories. Many PRM variants propose different strategies for each of the steps and choosing among them is problem dependent. Planning in heterogeneous environments and/or on parallel machines necessitates dividing the problem into regions where these choices have to be made for each one. Hand-selecting the best method for each region becomes infeasible. In particular, there are many ways to select connection candidates, and choosing the appropriate strategy is input dependent. In this paper, we present a general connection framework that adaptively selects a neighbor finding strategy from a candidate set of options. Our framework learns which strategy to use by examining their success rates and costs. It frees the user of the burden of selecting the best strategy and allows the selection to change over time. We perform experiments on rigid bodies of varying geometry and articulated linkages up to 37 degrees of freedom. Our results show that strategy performance is indeed problem/region dependent, and our adaptive method harnesses their strengths. Over all problems studied, our method differs the least from manual selection of the best method, and if one were to manually select a single method across all problems, the performance can be quite poor. Our method is able to adapt to changing sampling density and learns different strategies for each region when the problem is partitioned for parallelism.