Mobile manipulation is a crucial problem in various real-world applications. However, existing methods have demonstrated unsatisfactory training efficiency and sparse rewards, requiring complex coordination strategies between the mobile base and arm. In this paper, we propose RM-Planner, a planning method for mobile manipulation tasks in unknown complex environments. By adopting a two-layer hierarchical framework, we utilize a whole-body Model Predictive Control (MPC)-based low-level planner to track subgoals and generate aggressive but safe joint commands throughout the entire manipulation process, while a Reinforcement Learning (RL)based high-level policy directly uses 3D point cloud representations of the environment, guiding the robot to achieve optimal manipulation postures based on current observations and specific task objectives. We conduct extensive simulations and real-world experiments, where RM-planner significantly outperforms state-of-the-art methods. Our code will be released at https://github.com/SYSU-RoboticsLab/RM-Planner.git.

Positional and structural encodings (PSE) enable better identifiability of nodes within a graph, rendering them essential tools for empowering modern GNNs, and in particular graph Transformers. However, designing PSEs that work optimally for all graph prediction tasks is a challenging and unsolved problem. Here, we present the Graph Positional and Structural Encoder (GPSE), the first-ever graph encoder designed to capture rich PSE representations for augmenting any GNN. GPSE learns an efficient common latent representation for multiple PSEs, and is highly transferable: The encoder trained on a particular graph dataset can be used effectively on datasets drawn from markedly different distributions and modalities. We show that across a wide range of benchmarks, GPSE-enhanced models can significantly outperform those that employ explicitly computed PSEs, and at least match their performance in others. Our results pave the way for the development of foundational pre-trained graph encoders for extracting positional and structural information, and highlight their potential as a more powerful and efficient alternative to explicitly computed PSEs and existing self-supervised pre-training approaches. Our framework and pre-trained models are publicly available at https://github.com/G-Taxonomy-Workgroup/GPSE. For convenience, GPSE has also been integrated into the PyG library to facilitate downstream applications.

Network-based machine learning (ML) has the potential for predicting novel genes associated with nearly any health and disease context. However, this approach often uses network information from only the single species under consideration even though networks for most species are noisy and incomplete. While some recent methods have begun addressing this shortcoming by using networks from more than one species, they lack one or more key desirable properties: handling networks from more than two species simultaneously, incorporating many-to-many orthology information, or generating a network representation that is reusable across different types of and newly-defined prediction tasks. Here, we present GenePlexusZoo, a framework that casts molecular networks from multiple species into a single reusable feature space for network-based ML. We demonstrate that this multi-species network representation improves both gene classification within a single species and knowledge-transfer across species, even in cases where the inter-species correspondence is undetectable based on shared orthologous genes. Thus, GenePlexusZoo enables effectively leveraging the high evolutionary molecular, functional, and phenotypic conservation across species to discover novel genes associated with diverse biological contexts.

Single-cell technologies are revolutionizing the entire field of biology. The large volumes of data generated by single-cell technologies are high dimensional, sparse, and heterogeneous and have complicated dependency structures, making analyses using conventional machine learning approaches challenging and impractical. In tackling these challenges, deep learning often demonstrates superior performance compared to traditional machine learning methods. In this work, we give a comprehensive survey on deep learning in single-cell analysis. We first introduce background on single-cell technologies and their development, as well as fundamental concepts of deep learning including the most popular deep architectures. We present an overview of the single-cell analytic pipeline pursued in research applications while noting divergences due to data sources or specific applications. We then review seven popular tasks spanning different stages of the single-cell analysis pipeline, including multimodal integration, imputation, clustering, spatial domain identification, cell-type deconvolution, cell segmentation, and cell-type annotation. Under each task, we describe the most recent developments in classical and deep learning methods and discuss their advantages and disadvantages. Deep learning tools and benchmark datasets are also summarized for each task. Finally, we discuss the future directions and the most recent challenges. This survey will serve as a reference for biologists and computer scientists, encouraging collaborations.

The recent development of multimodal single-cell technology has made the possibility of acquiring multiple omics data from individual cells, thereby enabling a deeper understanding of cellular states and dynamics. Nevertheless, the proliferation of multimodal single-cell data also introduces tremendous challenges in modeling the complex interactions among different modalities. The recently advanced methods focus on constructing static interaction graphs and applying graph neural networks (GNNs) to learn from multimodal data. However, such static graphs can be suboptimal as they do not take advantage of the downstream task information; meanwhile GNNs also have some inherent limitations when deeply stacking GNN layers. To tackle these issues, in this work, we investigate how to leverage transformers for multimodal single-cell data in an end-to-end manner while exploiting downstream task information. In particular, we propose a scMoFormer framework which can readily incorporate external domain knowledge and model the interactions within each modality and cross modalities. Extensive experiments demonstrate that scMoFormer achieves superior performance on various benchmark datasets. Remarkably, scMoFormer won a Kaggle silver medal with the rank of 24/1221 (Top 2%) without ensemble in a NeurIPS 2022 competition1. Our implementation is publicly available at Github2.

MOTIVATION: Accurately representing biological networks in a low-dimensional space, also known as network embedding, is a critical step in network-based machine learning and is carried out widely using node2vec, an unsupervised method based on biased random walks. However, while many networks, including functional gene interaction networks, are dense, weighted graphs, node2vec is fundamentally limited in its ability to use edge weights during the biased random walk generation process, thus under-using all the information in the network. RESULTS: Here, we present node2vec+, a natural extension of node2vec that accounts for edge weights when calculating walk biases and reduces to node2vec in the cases of unweighted graphs or unbiased walks. Using two synthetic datasets, we empirically show that node2vec+ is more robust to additive noise than node2vec in weighted graphs. Then, using genome-scale functional gene networks to solve a wide range of gene function and disease prediction tasks, we demonstrate the superior performance of node2vec+ over node2vec in the case of weighted graphs. Notably, due to the limited amount of training data in the gene classification tasks, graph neural networks such as GCN and GraphSAGE are outperformed by both node2vec and node2vec+. AVAILABILITY AND IMPLEMENTATION: The data and code are available on GitHub at https://github.com/krishnanlab/node2vecplus_benchmarks. All additional data underlying this article are available on Zenodo at https://doi.org/10.5281/zenodo.7007164. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: PyGenePlexus is a Python package that enables a user to gain insight into any gene set of interest through a molecular interaction network informed supervised machine learning model. PyGenePlexus provides predictions of how associated every gene in the network is to the input gene set, offers interpretability by comparing the model trained on the input gene set to models trained on thousands of known gene sets, and returns the network connectivity of the top predicted genes. AVAILABILITY AND IMPLEMENTATION: https://pypi.org/project/geneplexus/ and https://github.com/krishnanlab/PyGenePlexus. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Learning low-dimensional representations (embeddings) of nodes in large graphs is key to applying machine learning on massive biological networks. Node2vec is the most widely used method for node embedding. However, its original Python and C++ implementations scale poorly with network density, failing for dense biological networks with hundreds of millions of edges. We have developed PecanPy, a new Python implementation of node2vec that uses cache-optimized compact graph data structures and precomputing/parallelization to result in fast, high-quality node embeddings for biological networks of all sizes and densities. AVAILABILITYAND IMPLEMENTATION: PecanPy software is freely available at https://github.com/krishnanlab/PecanPy. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Assigning every human gene to specific functions, diseases and traits is a grand challenge in modern genetics. Key to addressing this challenge are computational methods, such as supervised learning and label propagation, that can leverage molecular interaction networks to predict gene attributes. In spite of being a popular machine-learning technique across fields, supervised learning has been applied only in a few network-based studies for predicting pathway-, phenotype- or disease-associated genes. It is unknown how supervised learning broadly performs across different networks and diverse gene classification tasks, and how it compares to label propagation, the widely benchmarked canonical approach for this problem. RESULTS: In this study, we present a comprehensive benchmarking of supervised learning for network-based gene classification, evaluating this approach and a classic label propagation technique on hundreds of diverse prediction tasks and multiple networks using stringent evaluation schemes. We demonstrate that supervised learning on a gene's full network connectivity outperforms label propagaton and achieves high prediction accuracy by efficiently capturing local network properties, rivaling label propagation's appeal for naturally using network topology. We further show that supervised learning on the full network is also superior to learning on node embeddings (derived using node2vec), an increasingly popular approach for concisely representing network connectivity. These results show that supervised learning is an accurate approach for prioritizing genes associated with diverse functions, diseases and traits and should be considered a staple of network-based gene classification workflows. AVAILABILITY AND IMPLEMENTATION: The datasets and the code used to reproduce the results and add new gene classification methods have been made freely available. CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Owing to environmental concerns, the energy-efficient subway train scheduling problem is necessary in subway operation management. This paper designs an approximate dynamic programming (DP) approach for energy-efficient subway train scheduling problem with time-dependent demand. The train traffic model is proposed with the dynamic equations for the evolution of train headway, train passenger loads, and the energy consumption along the subway line. For the dynamic changing of the onboard passengers with time, the total train energy usage is modeled as the sum of energy consumptions from the traction system and auxiliary facilities. A nonlinear DP problem is formulated to generate a near optimal timetable to realize the tradeoff among the utilization of trains, passenger waiting time, service levels, and energy consumption. To overcome the curse of dimensionality in this optimization problem, we construct an approximate DP framework, where the conceptions of states, policies, state transitions, and reward function are introduced. And this algorithm is able to converge to a good solution with a short time compared to the genetic algorithm and differential evolution algorithm. Finally, the numerical experiments are given to demonstrate the effectiveness of the proposed model and algorithm.