• scaLR: a low-resource deep neu...

Treffer: scaLR: a low-resource deep neural network-based platform for single cell analysis and biomarker discovery.

Title:
scaLR: a low-resource deep neural network-based platform for single cell analysis and biomarker discovery.
Authors:
Jogani S; Department of Generative AI & Bioinformatics, Infocusp Innovations, Laxman Nagar Baner, Pune 411045, Maharashtra, India., Pol AS; Department of Generative AI & Bioinformatics, Infocusp Innovations, Laxman Nagar Baner, Pune 411045, Maharashtra, India., Prajapati M; Department of Generative AI & Bioinformatics, Infocusp Innovations, Gala-hub, Bopal, Ahmedabad 380058, Gujarat, India., Samal A; Department of Generative AI & Bioinformatics, Infocusp Innovations, Gala-hub, Bopal, Ahmedabad 380058, Gujarat, India., Bhatia K; Department of Generative AI & Bioinformatics, Infocusp Innovations, Laxman Nagar Baner, Pune 411045, Maharashtra, India., Parmar J; Department of Generative AI & Bioinformatics, Infocusp Innovations, Gala-hub, Bopal, Ahmedabad 380058, Gujarat, India., Patel U; Department of Generative AI & Bioinformatics, Infocusp Innovations, Gala-hub, Bopal, Ahmedabad 380058, Gujarat, India., Shah F; Department of Generative AI & Bioinformatics, Infocusp Innovations, Gala-hub, Bopal, Ahmedabad 380058, Gujarat, India., Vyas N; Department of Generative AI & Bioinformatics, Infocusp Innovations, Gala-hub, Bopal, Ahmedabad 380058, Gujarat, India., Gupta S; Department of Generative AI & Bioinformatics, Infocusp Innovations, Gala-hub, Bopal, Ahmedabad 380058, Gujarat, India.
Source:
Briefings in bioinformatics [Brief Bioinform] 2025 May 01; Vol. 26 (3).
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: Oxford University Press Country of Publication: England NLM ID: 100912837 Publication Model: Print Cited Medium: Internet ISSN: 1477-4054 (Electronic) Linking ISSN: 14675463 NLM ISO Abbreviation: Brief Bioinform Subsets: MEDLINE
Imprint Name(s):
Publication: Oxford : Oxford University Press
Original Publication: London ; Birmingham, AL : H. Stewart Publications, [2000-
MeSH Terms:
Single-Cell Analysis*/methods , Neural Networks, Computer* , Biomarkers*/metabolism , Software* , Computational Biology*/methods , Deep Learning*, Humans ; Algorithms ; Sequence Analysis, RNA
References:
iScience. 2021 Oct 22;24(10):103115. (PMID: 34522848)
Nat Commun. 2022 Mar 10;13(1):1246. (PMID: 35273156)
Front Genet. 2020 May 12;11:490. (PMID: 32477414)
Commun Biol. 2023 May 20;6(1):545. (PMID: 37210444)
Nat Methods. 2018 May;15(5):359-362. (PMID: 29608555)
Nat Genet. 2021 Aug;53(8):1143-1155. (PMID: 34239132)
Nat Commun. 2022 Sep 7;13(1):5271. (PMID: 36071107)
Proc Natl Acad Sci U S A. 2020 Oct 13;117(41):25800-25809. (PMID: 32989152)
Mol Syst Biol. 2021 Aug;17(8):e10240. (PMID: 34432947)
Front Mol Neurosci. 2015 Dec 16;8:75. (PMID: 26733799)
Genome Biol. 2021 Feb 22;22(1):69. (PMID: 33618746)
Nat Immunol. 2019 Feb;20(2):163-172. (PMID: 30643263)
Nat Biotechnol. 2022 Feb;40(2):163-166. (PMID: 35132262)
Nat Neurosci. 2021 Feb;24(2):276-287. (PMID: 33432193)
Bioinformatics. 2022 Feb 7;38(5):1393-1402. (PMID: 34893819)
Brief Bioinform. 2023 Sep 20;24(5):. (PMID: 37523217)
Nat Med. 2020 Mar;26(3):333-340. (PMID: 32066974)
Biomedicines. 2021 Oct 21;9(11):. (PMID: 34829742)
Cell Syst. 2019 Aug 28;9(2):207-213.e2. (PMID: 31377170)
Immunity. 2021 Apr 13;54(4):797-814.e6. (PMID: 33765436)
Immunobiology. 2021 May;226(3):152092. (PMID: 34004549)
Mol Genet Genomic Med. 2020 Sep;8(9):e1373. (PMID: 32588496)
Biomed Pharmacother. 2023 Sep;165:115077. (PMID: 37393865)
Bioinformatics. 2020 Jan 15;36(2):533-538. (PMID: 31359028)
Neuron. 2022 Oct 05;110(19):3139-3153.e6. (PMID: 35998632)
Neuron. 2022 Sep 21;110(18):2929-2948.e8. (PMID: 35882228)
Brief Bioinform. 2022 Mar 10;23(2):. (PMID: 35043143)
Nat Genet. 2024 Apr;56(4):652-662. (PMID: 38548988)
Acta Biochim Pol. 2025 Feb 05;72:13922. (PMID: 39980637)
Genomics Proteomics Bioinformatics. 2022 Oct;20(5):814-835. (PMID: 36528240)
Nature. 2022 Mar;603(7903):885-892. (PMID: 35165441)
Genome Biol. 2019 Sep 9;20(1):194. (PMID: 31500660)
IEEE J Biomed Health Inform. 2024 May;28(5):3134-3145. (PMID: 38709615)
IEEE/ACM Trans Comput Biol Bioinform. 2023 Nov-Dec;20(6):3535-3546. (PMID: 37486829)
Genome Biol. 2020 Jan 16;21(1):12. (PMID: 31948481)
Bioinformatics. 2019 Nov 1;35(22):4688-4695. (PMID: 31028376)
iScience. 2020 Mar 27;23(3):100882. (PMID: 32062421)
Brief Bioinform. 2022 Sep 20;23(5):. (PMID: 35914950)
Nat Biotechnol. 2018 Jun;36(5):421-427. (PMID: 29608177)
Cereb Cortex. 2007 Jul;17(7):1712-21. (PMID: 17021275)
Nat Commun. 2021 Oct 6;12(1):5849. (PMID: 34615861)
Nat Biotechnol. 2024 Feb;42(2):293-304. (PMID: 37231261)
Brain Behav. 2024 Feb;14(2):e3401. (PMID: 39470400)
Nature. 2022 Aug;608(7924):766-777. (PMID: 35948637)
Clin Genet. 2022 Nov;102(5):379-390. (PMID: 35882622)
Bioinformatics. 2021 Oct 11;37(19):3228-3234. (PMID: 33904573)
Genomics Proteomics Bioinformatics. 2021 Apr;19(2):267-281. (PMID: 33359678)
Nucleic Acids Res. 2025 Jan 6;53(D1):D886-D900. (PMID: 39607691)
Nat Methods. 2024 Aug;21(8):1470-1480. (PMID: 38409223)
Bioinformatics. 2023 Sep 2;39(9):. (PMID: 37669147)
Science. 2022 Apr 8;376(6589):eabf1970. (PMID: 35389781)
Comput Struct Biotechnol J. 2021 Jan 19;19:961-969. (PMID: 33613863)
Nucleic Acids Res. 2019 Sep 19;47(16):e95. (PMID: 31226206)
Front Mol Neurosci. 2022 Mar 08;15:848506. (PMID: 35350431)
Caspian J Intern Med. 2013 Spring;4(2):627-35. (PMID: 24009950)
Brief Bioinform. 2020 Sep 25;21(5):1581-1595. (PMID: 31675098)
Nature. 2024 Feb;626(8001):1084-1093. (PMID: 38355799)
Nat Methods. 2019 Oct;16(10):983-986. (PMID: 31501545)
Science. 2022 May 13;376(6594):eabl5197. (PMID: 35549406)
HGG Adv. 2023 Oct 12;4(4):100236. (PMID: 37660254)
Science. 2020 Nov 13;370(6518):. (PMID: 33184181)
Blood Adv. 2024 Jul 23;8(14):3665-3678. (PMID: 38507736)
Nucleic Acids Res. 2023 Jan 6;51(D1):D870-D876. (PMID: 36300619)
Cell. 2022 Mar 3;185(5):916-938.e58. (PMID: 35216673)
Genes (Basel). 2019 Jul 12;10(7):. (PMID: 31336988)
Genome Biol. 2021 Sep 9;22(1):264. (PMID: 34503564)
Nat Methods. 2019 Oct;16(10):1007-1015. (PMID: 31501550)
Grant Information:
Infocusp Innovation Research funds
Contributed Indexing:
Keywords: ML/DNN; annotation; cell type; classification; feature selection
Substance Nomenclature:
0 (Biomarkers)
Entry Date(s):
Date Created: 20250529 Date Completed: 20250529 Latest Revision: 20250601
Update Code:
20250601
PubMed Central ID:
PMC12121358
DOI:
10.1093/bib/bbaf243
PMID:
40439670
Database:
MEDLINE

Weitere Informationen

Single-cell ribonucleic acid (RNA) sequencing (scRNA-seq) produces vast amounts of individual cell profiling data. Its analysis presents a significant challenge in accurately annotating cell types and their associated biomarkers. Different pipelines based on deep neural network (DNN) methods have been employed to tackle these issues. These pipelines have arisen as a promising resource and can extract meaningful and concise features from noisy, diverse, and high-dimensional data to enhance annotations and subsequent analysis. Existing tools require high computational resources to execute large sample datasets. We have developed a cutting-edge platform known as scaLR (Single-cell analysis using low resource) that efficiently processes data into feature subsets, samples in batches to reduce the required memory for processing large datasets, and running DNN models in multiple central processing units. scaLR is equipped with data processing, feature extraction, training, evaluation, and downstream analysis. Its novel feature extraction algorithm first trains the model on a feature subset and stores the importance of the features for all the features in that subset. At the end of the training of all subsets, the top-K features are selected based on their importance. The final model is trained on top-K features; its performance evaluation and associated downstream analysis provide significant biomarkers for different cell types and diseases/traits. Our findings indicate that scaLR offers comparable prediction accuracy and requires less model training time and computational resources than existing Python-based pipelines. We present scaLR, a Python-based platform, engineered to utilize minimal computational resources while maintaining comparable execution times and analysis costs to existing frameworks.
(© The Author(s) 2025. Published by Oxford University Press.)