A novel deep autoencoder architecture is proposed for the analysis of histopathology images. Its purpose is to produce a disentangled latent representation in which the structure and colour information are confined to different subspaces so that stain-independent models may be learned. For this, we introduce two constraints on the representation which are implemented as a classifier and an adversarial discriminator. We show how they can be used for learning a latent representation across haematoxylin-eosin and a number of immune stains. Finally, we demonstrate the utility of the proposed representation in the context of matching image patches for registration applications and for learning a bag of visual words for whole slide image summarization.
MSMetaEnhancer is a Python software package for the metadata enrichment of records in mass spectral library files commonly used as reference for chemical identification via mass spectrometry. Each record contains spectral information, i.e., peak mass to charge ratio (m/z) and intensities, alongside chemical & structural metadata, e.g., identifiers. The package uses matchms (Huber et al., 2020) for data IO and supports the open, text-based .msp format. It annotates given mass spectra records in the library file by adding missing metadata such as SMILES, InChI, and CAS numbers to the individual entries. The package retrieves the respective information by querying several external databases using existing metadata (e.g., SMILES or CAS number), converting different representations or database identifiers. Multiple databases and services are included, currently supporting the chemical identifier resolver (CIR), chemical translation service (CTS) (Wohlgemuth et al., 2010), ChemIDplus (Tomasulo, 2002, the Integrated Database for Small Molecules (IDSM) (Galgonek & Vondrášek, 2021), PubChem (Kim et al., 2021), and BridgeDb (van Iersel et al., 2010. Additionally, instead of querying external databases, computing the identifiers is also supported (e.g. molecular weight from SMILES).
The Architecture, Engineering and Construction (AEC) is in its transition from 2D design processes to 3D object-oriented modelling. Building Information Modeling (BIM) is steadily gaining importance, replacing the conventional Computer-Aided Design (CAD) practices and getting implemented in every aspect of the very complex software and stakeholder landscape (Jaud et al., 2019). As one of the main principles, BIM describes the idea of integrating all information relevant to the life cycle of a structure, such as a tunnel, bridge, building or road, in a digital (BIM) model. The digital model is to ensure, among other things (Amann, 2018):• that all relevant data is available to all project participants; • that all data is in a consistent state (data integrity should be guaranteed); and • that the data can be used efficiently. TUM Open Infra Platform (OIP) is an open source application for viewing and analysis of different BIM models used in the civil engineering field. OIP supports reading, visualization, navigating, and handling of:
RIAssigner is a software package for the computation of gas chromatographic (GC) retention indices (RI). The package uses matchms (Huber et al., 2020) and pandas (The pandas development team, 2020) for data I/O and supports the .msp as well as tabular (.csv & .tsv) formats, among others. It supports multiple keywords identifying the retention time (RT) and RI information and handling SI units for RT. The RI can be computed using non-isothermal Kováts retention-indexing (from temperature programming, using the definition of van Den Dool & Kratz (1963)) or cubic spline interpolation (Halang et al., 1978) based on a reference dataset containing RT & RI. The MIT-licensed package is hosted via bioconda (Grüning et al., 2018) and is also accessible to users as a Galaxy tool (Jalili et al., 2020; Spectrometric Data Processing and Analysis & Institute of Computer Science, 2022).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.