PubChem (https://pubchem.ncbi.nlm.nih.gov) is a key chemical information resource for the biomedical research community. Substantial improvements were made in the past few years. New data content was added, including spectral information, scientific articles mentioning chemicals, and information for food and agricultural chemicals. PubChem released new web interfaces, such as PubChem Target View page, Sources page, Bioactivity dyad pages and Patent View page. PubChem also released a major update to PubChem Widgets and introduced a new programmatic access interface, called PUG-View. This paper describes these new developments in PubChem.

/pdf/pubchem-2019-update-improved-access-to-chemical-data-140jmv6hsa.pdf

PubChem 2019 update: improved access to chemical data

Provided herein are methods and systems relating to location-based services such as social networking, providing demographic information, tracking mobile devices, providing business information, providing an adaptable user interface, remotely effecting a change on a portable electronic device, providing a geofence, outputting location-based information on a mobile device, varying transmissions to and from a mobile device, providing location-based alerts, verifying transactions and tailoring information to the behavior of a user.

Location-based services

Numerous important problems can be framed as learning from graph data. We propose a framework for learning convolutional neural networks for arbitrary graphs. These graphs may be undirected, directed, and with both discrete and continuous node and edge attributes. Analogous to image-based convolutional networks that operate on locally connected regions of the input, we present a general approach to extracting locally connected regions from graphs. Using established benchmark data sets, we demonstrate that the learned feature representations are competitive with state of the art graph kernels and that their computation is highly efficient.

Learning Convolutional Neural Networks for Graphs

Deep learning describes a class of machine learning algorithms that are capable of combining raw inputs into layers of intermediate features. These algorithms have recently shown impressive results across a variety of domains. Biology and medicine are data-rich disciplines, but the data are complex and often ill-understood. Hence, deep learning techniques may be particularly well suited to solve problems of these fields. We examine applications of deep learning to a variety of biomedical problems-patient classification, fundamental biological processes and treatment of patients-and discuss whether deep learning will be able to transform these tasks or if the biomedical sphere poses unique challenges. Following from an extensive literature review, we find that deep learning has yet to revolutionize biomedicine or definitively resolve any of the most pressing challenges in the field, but promising advances have been made on the prior state of the art. Even though improvements over previous baselines have been modest in general, the recent progress indicates that deep learning methods will provide valuable means for speeding up or aiding human investigation. Though progress has been made linking a specific neural network's prediction to input features, understanding how users should interpret these models to make testable hypotheses about the system under study remains an open challenge. Furthermore, the limited amount of labelled data for training presents problems in some domains, as do legal and privacy constraints on work with sensitive health records. Nonetheless, we foresee deep learning enabling changes at both bench and bedside with the potential to transform several areas of biology and medicine.

https://royalsocietypublishing.org/doi/pdf/10.1098/rsif.2017.0387

Opportunities and obstacles for deep learning in biology and medicine.

Hypertension 66 (20.3%) 24 (24.2%) 30 (16.3%) NS Diabetes 20 (6.2%) 7 (7.1%) 10 (5.4%) NS Excess weight 78 (24%) 27 (27.3%) 44 (23.9%) NS Smokers 64 (19.7%) 17 (17.2%) 35 (19.0%) NS Age >50 years 137 (42.2%) 54 (54.5%) 67 (36.4%) <0.02 Kidney disease 7 (2.2%) 1 (1%) 5 (2.7%) NS Family history, DM 102 (31.4%) 28 (28.3%) 66 (35.9%) NS

Conflict of interest statement. None declared.

Deep convolutional neural networks comprise a subclass of deep neural networks (DNN) with a constrained architecture that leverages the spatial and temporal structure of the domain they model. Convolutional networks achieve the best predictive performance in areas such as speech and image recognition by hierarchically composing simple local features into complex models. Although DNNs have been used in drug discovery for QSAR and ligand-based bioactivity predictions, none of these models have benefited from this powerful convolutional architecture. This paper introduces AtomNet, the first structure-based, deep convolutional neural network designed to predict the bioactivity of small molecules for drug discovery applications. We demonstrate how to apply the convolutional concepts of feature locality and hierarchical composition to the modeling of bioactivity and chemical interactions. In further contrast to existing DNN techniques, we show that AtomNet's application of local convolutional filters to structural target information successfully predicts new active molecules for targets with no previously known modulators. Finally, we show that AtomNet outperforms previous docking approaches on a diverse set of benchmarks by a large margin, achieving an AUC greater than 0.9 on 57.8% of the targets in the DUDE benchmark.

AtomNet: A Deep Convolutional Neural Network for Bioactivity Prediction in Structure-based Drug Discovery

Undetected overfitting can occur when there are significant redundancies between training and validation data. We describe AVE, a new measure of training–validation redundancy for ligand-based classification problems, that accounts for the similarity among inactive molecules as well as active ones. We investigated seven widely used benchmarks for virtual screening and classification, and we show that the amount of AVE bias strongly correlates with the performance of ligand-based predictive methods irrespective of the predicted property, chemical fingerprint, similarity measure, or previously applied unbiasing techniques. Therefore, it may be the case that the previously reported performance of most ligand-based methods can be explained by overfitting to benchmarks rather than good prospective accuracy.

Most Ligand-Based Classification Benchmarks Reward Memorization Rather than Generalization.

The present invention is a system and method for handling location information. The goal of the present invention is to allow users to collect and evaluate information from multiple sources, and thus make proper use of all available location information. Location sources for the present invention include mobile electronic devices such as mobile telephones, cell phones, hand-held computers, personal digital assistants, pagers, Global Positioning System (GPS) devices, and other pervasive computing devices. Satellite-based or network-based positioning technologies make it possible to determine the geographic location of these wireless electronic devices and their users (e.g. location-based services for users of mobile telephones use such positioning technologies). Location sources for the present invention also include computer-based models, schedules or calendars that give a person's expected location depending on the date and time. One aspect of the present invention is a method for handling location information. Another aspect of the present invention is a system for executing the method of the present invention. A third aspect of the present invention is as a set of instructions on a computer-usable medium, or resident in a computer system, for executing the method of the present invention.

System and method for handling location information

This paper describes an experimental system in which customized high performance XML parsers are prepared using parser generation and compilation techniques. Parsing is integrated with Schema-based validation and deserialization, and the resulting validating processors are shown to be as fast as or in many cases significantly faster than traditional nonvalidating parsers. High performance is achieved by integration across layers of software that are traditionally separate, by avoiding unnecessary data copying and transformation, and by careful attention to detail in the generated code. The effect of API design on XML performance is also briefly discussed..

/pdf/xml-screamer-an-integrated-approach-to-high-performance-xml-4r6g3hvxhn.pdf

XML screamer: an integrated approach to high performance XML parsing, validation and deserialization

Provided is a system and method that acquires and aggregates information to determine the location of a user. Information from multiple sources is collected and evaluated. Location sources include mobile electronic devices such as mobile telephones, cell phones, hand-held computers, personal digital assistants, pagers, Global Positioning System (GPS) devices, and other pervasive computing devices. Satellite-based or network-based positioning technologies make it possible to determine the geographic location of these wireless electronic devices and their corresponding user. Location sources also include computer-based models, schedules or calendars that give a particular user's expected location depending on the date and time. Provided is a method for aggregating information to determine a user's location. Also provided is a system for executing the claimed method. Also provided is as a set of instructions on a computer-usable medium, or resident in a computer system, for executing the claimed method.

Abraham Heifets

Papers

AtomNet: A Deep Convolutional Neural Network for Bioactivity Prediction in Structure-based Drug Discovery

Most Ligand-Based Classification Benchmarks Reward Memorization Rather than Generalization.

System and method for handling location information

XML screamer: an integrated approach to high performance XML parsing, validation and deserialization

System and method for aggregating information to determine users' locations