Sanitary landfill site selection using GIS-based on a fuzzy multi-criteria evaluation technique: a case study of the City of Kraljevo, Serbia.

Adequate disposal of municipal solid waste (MSW) is one of Serbia's most complex environmental challenges. The problem is more serious in urban areas, since large amounts of waste are disposed of in locations that do not comply with environmental, technical, and socio-economic standards. Such is the case for the city of Kraljevo, where about 116,000 inhabitants do not have a sanitary landfill facility. This research includes a multi-criteria analysis, conducted with the help of geographic information systems, to find a suitable landfill site location. After data collection, the first step was to process 15 environmental and socio-economic factors utilizing the fuzzy analytic-hierarchy process method. The second step comprised the visual analysis and selection of the ten most suitable locations from the synthetic convenience map. The third step involved the final ranking of sites by means of the fuzzy multi-objective analysis by ratio, plus the full multiplicative form method, based on four additional beneficial and non-beneficial criteria. The results show that sanitary landfill candidate site A4 is the most suitable location for constructing a sanitary landfill site due to its large area (569 ha) and relatively short distance from the urban zone (8 km). This study is the first to integrate geographic information systems and the fuzzy analytic-hierarchy process, multi-objective analysis by ratio, and the full multiplicative form algorithm for sanitary landfill selection. The results of the research can be used as a reference for safe waste disposal in the city of Kraljevo.

A community-powered search of machine learning strategy space to find NMR property prediction models.

The rise of machine learning (ML) has created an explosion in the potential strategies for using data to make scientific predictions. For physical scientists wishing to apply ML strategies to a particular domain, it can be difficult to assess in advance what strategy to adopt within a vast space of possibilities. Here we outline the results of an online community-powered effort to swarm search the space of ML strategies and develop algorithms for predicting atomic-pairwise nuclear magnetic resonance (NMR) properties in molecules. Using an open-source dataset, we worked with Kaggle to design and host a 3-month competition which received 47,800 ML model predictions from 2,700 teams in 84 countries. Within 3 weeks, the Kaggle community produced models with comparable accuracy to our best previously published 'in-house' efforts. A meta-ensemble model constructed as a linear combination of the top predictions has a prediction accuracy which exceeds that of any individual model, 7-19x better than our previous state-of-the-art. The results highlight the potential of transformer architectures for predicting quantum mechanical (QM) molecular properties.

Improved Scaffold Hopping in Ligand-Based Virtual Screening Using Neural Representation Learning.

Deep learning has demonstrated significant potential in advancing state of the art in many problem domains, especially those benefiting from automated feature extraction. Yet, the methodology has seen limited adoption in the field of ligand-based virtual screening (LBVS) as traditional approaches typically require large, target-specific training sets, which limits their value in most prospective applications. Here, we report the development of a neural network architecture and a learning framework designed to yield a generally applicable tool for LBVS. Our approach uses the molecular graph as input and involves learning a representation that places compounds of similar biological profiles in close proximity within a hyperdimensional feature space; this is achieved by simultaneously leveraging historical screening data against a multitude of targets during training. Cosine distance between molecules in this space becomes a general similarity metric and can readily be used to rank order database compounds in LBVS workflows. We demonstrate the resulting model generalizes exceptionally well to compounds and targets not used in its training. In three commonly employed LBVS benchmarks, our method outperforms popular fingerprinting algorithms without the need for any target-specific training. Moreover, we show the learned representation yields superior performance in scaffold hopping tasks and is largely orthogonal to existing fingerprints. Summarily, we have developed and validated a framework for learning a molecular representation that is applicable to LBVS in a target-agnostic fashion, with as few as one query compound. Our approach can also enable organizations to generate additional value from large screening data repositories, and to this end we are making its implementation freely available at https://github.com/totient-bio/gatnn-vs.