SAVI-Space

DOI

This dataset was published as part of SAVI Space—combinatorial encoding of the billion-size synthetically accessible virtual inventory (Korn, M., Judson, P., Klein, R. et al.,Sci Data (12) 1064, 2025) and holds the combinatorial chemical fragment space of the Synthetically Accessible Virtual Inventory (SAVI) Database. The combinatorial fragment space is optimized for fast fingerprint similarity search with SpaceLight (Bellmann et al. J. Chem. Inf. 460 Model. 61 2020), and for fast maximum-common-substructure search with SpaceMACS (Schmidt et al., J. Chem. Inf. Model. 62 (9) 2133–2150, 2022), available at https://software.zbh.uni-hamburg.de.

Besides the chemical fragment space mimic the SAVI-Lib-2020 (SAVI-Space-2020-Librules) there is a updated version based on adapted rules called SAVI-Space-2020, and the Enamine Building Blocks (June 2024) called SAVI-Space-2024. Additional, the preprocessed building blocks of the SAVI-Space-2020-Librules and SAVI-Space-2020 are available.

The SAVI-Space-2020(SAVI-Lib-Rules) is available as Database file (SAVI-Space-2020-Librules.tfsdb) and can be opened with SpaceLight Version 1.2.3 available at https://software.zbh.uni-hamburg.de.

The SAVI-Space-2020 and SAVI-Space-2024 are available as space files (SAVI-Space-{2020,2024}.space) and can be opened with the SpaceLightN Version 1.3.0 and SpaceMACS Version 1.1.0 also available at https://software.zbh.uni-hamburg.de.

Note: SpaceLight and SpaceLightN refer to different versions of the tool.

ℹ Info: This version of SAVI-Space isn’t fully optimized for SpaceMACS and needs more RAM than usual. Please allocate at least 64 GB for a successful run.

Identifier
DOI https://doi.org/10.25592/uhhfdm.15990
Related Identifier IsNewVersionOf https://doi.org/10.1038/s41597-020-00727-4
Related Identifier IsNewVersionOf https://doi.org/10.35115/37N9-5738
Related Identifier IsReferencedBy https://doi.org/10.1038/s41597-025-05384-z
Related Identifier IsPartOf https://doi.org/10.25592/uhhfdm.15989
Metadata Access https://www.fdr.uni-hamburg.de/oai2d?verb=GetRecord&metadataPrefix=oai_datacite&identifier=oai:fdr.uni-hamburg.de:15990
Provenance
Creator Korn, Malte ORCID logo
Publisher Universität Hamburg
Contributor Klein, Raphael; Judson, Philip; Lemmen, Christian; Nicklaus, Marc C.; Rarey, Matthias
Publication Year 2025
Rights Creative Commons Attribution 4.0 International; Open Access; https://creativecommons.org/licenses/by/4.0/legalcode; info:eu-repo/semantics/openAccess
OpenAccess true
Contact Rarey, Matthias (University of Hamburg)
Representation
Resource Type Dataset
Version 1.0.0
Discipline Other