Efficiently accelerated bioimage analysis with NanoPyx, a Liquid Engine-powered Python framework

: Saraiva, Bruno M.; Cunha, Inês; Brito, António D.; Follain, Gautier; Portela, Raquel; Haase, Robert; Pereira, Pedro M.; Jacquemet, Guillaume; Henriques, Ricardo

Publisher: Springer Nature

: 2025

: Nature Methods

: 22

: 283

: 286

: 1548-7091

: 1548-7105

DOI: https://doi.org/10.1038/s41592-024-02562-6

: https://doi.org/10.1038/s41592-024-02562-6

: https://research.utu.fi/converis/portal/detail/Publication/484850987

The expanding scale and complexity of microscopy image datasets require accelerated analytical workflows. NanoPyx meets this need through an adaptive framework enhanced for high-speed analysis. At the core of NanoPyx, the Liquid Engine dynamically generates optimized central processing unit and graphics processing unit code variations, learning and predicting the fastest based on input data and hardware. This data-driven optimization achieves considerably faster processing, becoming broadly relevant to reactive microscopy and computing fields requiring efficiency.

s41592-024-02562-6.pdf

:
R. Henriques, P.M.P. and R.P. acknowledge support from LS4FUTURE Associated Laboratory (no. LA/P/0087/2020). R. Henriques, B.M.S. and I.C. acknowledge the support of the Gulbenkian Foundation (Fundação Calouste Gulbenkian); the European Research Council under the European Union’s Horizon 2020 research and innovation program (grant agreement no. 101001332); the European Commission through the Horizon Europe program (AI4LIFE project with grant agreement no. 101057970-AI4LIFE and RT-SuperES project with grant agreement no. 101099654-RT-SuperES); the European Molecular Biology Organization Installation Grant (no. EMBO-2020-IG-4734); and the Chan Zuckerberg Initiative Visual Proteomics Grant (no. vpi-0000000044; https://doi.org/10.37921/743590vtudfp). In addition, A.D.B. acknowledges the FCT 2021.06849.BD fellowship. R. Henriques and B.M.S. also acknowledge that this project has been made possible in part by a grant from the Chan Zuckerberg Initiative DAF, an advised fund of Silicon Valley Community Foundations (Chan Zuckerberg Initiative Napari Plugin Foundations Grants Cycle 2, no. NP2-0000000085). P.M.P. and R.P. acknowledge support from Fundação para a Ciência e Tecnologia (Portugal) project grant no. PTDC/BIA-MIC/2422/2020 and the MOSTMICRO-ITQB R&D Unit (nos. UIDB/04612/2020 and UIDP/04612/2020). P.M.P. acknowledges support from La Caixa Junior Leader Fellowship (no. LCF/BQ/PI20/11760012), financed by ‘la Caixa’ Foundation (ID 100010434) and the European Union’s Horizon 2020 research and innovation program under Marie Skłodowska-Curie grant agreement no. 847648, and from a Maratona da Saúde award. This study was supported by the Academy of Finland (no. 338537 to G.J.), the Sigrid Juselius Foundation (to G.J.), the Cancer Society of Finland (Syöpäjärjestöt, to G.J.) and the Solutions for Health strategic funding to Åbo Akademi University (to G.J.). This research was supported by InFLAMES Flagship Program of the Academy of Finland (decision no. 337531).