Quantitation of metabolic activity from isotope tracing data using automated methodology

IF 18.9 1区 医学 Q1 ENDOCRINOLOGY & METABOLISM
Shiyu Liu, Xiaojing Liu, Jason W. Locasale
{"title":"Quantitation of metabolic activity from isotope tracing data using automated methodology","authors":"Shiyu Liu, Xiaojing Liu, Jason W. Locasale","doi":"10.1038/s42255-024-01144-2","DOIUrl":null,"url":null,"abstract":"<p>Metabolic flux analysis (MFA) is a computational approach to deciphering labelling patterns based on machine learning principles. Differing from typical machine learning algorithms that train a model from known datasets to make predictions, the commonly used MFA algorithm trains a metabolic network with data from isotope tracing experiments and directly outputs the learned information — that is, all fluxes in the network that best fit data<sup>3,5</sup> (Fig. 1b). However, as a machine learning algorithm, current MFA methods often lack systematic evaluation and benchmarking, a standard practice in broader machine learning and artificial intelligence applications<sup>6</sup>. Issues such as algorithmic convergence, flux estimation accuracy and result robustness in MFA studies have been raised but remain largely unaddressed<sup>3</sup>, limiting the effectiveness and broader adoption of these automated tools in metabolic research.</p><p>To advance the capabilities of MFA for complex metabolic networks and extensive isotope tracing datasets, we developed an automated analysis methodology alongside a large-scale metabolic network model. This model comprises over 100 fluxes across key pathways, including glycolysis, the tricarboxylic acid (TCA) cycle, the pentose phosphate pathway (PPP), one-carbon metabolism, and several amino acid (AA) biosynthetic pathways (Fig. 1c, Supplementary Methods). Compared to contemporary MFA tools<sup>7,8,9</sup>, a notable feature of our methodology is the incorporation of organelle compartmentalization, facilitating accurate quantification of exchange fluxes between mitochondria and cytosol in eukaryotic cells (Fig. 1c). While other tools typically require tens of minutes to obtain a solution<sup>7,9</sup>, our methodology can generate an optimized solution, with fluxes that accurately explain the labelling pattern from a <sup>13</sup>C tracing experiment on cultured cell lines, within 2 s on a desktop computer<sup>10</sup> (Supplementary Fig. 1a–d). Nonetheless, a challenge arose from the observation that these optimized solutions could diverge significantly, showing considerable variability in certain net fluxes even with similar loss values (Fig. 1d, Supplementary Fig. 1e,f).To address this problem, we developed an optimization-averaging algorithm that refines the computation process by selecting a subset of solutions with minimal loss (selected solutions) from the pool of optimized solutions and averaging them to produce a new, more stable solution set (averaged solutions) (Fig. 1e, Supplementary Methods). These solutions, along with those generated using the typical strategy used in contemporary software (Supplementary Methods, Supplementary Fig. 1c), were benchmarked using simulated <sup>13</sup>C tracing datasets generated from a known flux vector (Supplementary Fig. 2a). The results demonstrated that, relative to the benchmark, the optimization-averaging algorithm effectively reduced flux variability and improved the accuracy of the results in approximating the known flux, even with varying levels of data availability (Fig. 1f, Supplementary Figs. 2b–e and 3a–e).</p>","PeriodicalId":19038,"journal":{"name":"Nature metabolism","volume":"5 1","pages":""},"PeriodicalIF":18.9000,"publicationDate":"2024-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Nature metabolism","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1038/s42255-024-01144-2","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENDOCRINOLOGY & METABOLISM","Score":null,"Total":0}
引用次数: 0

Abstract

Metabolic flux analysis (MFA) is a computational approach to deciphering labelling patterns based on machine learning principles. Differing from typical machine learning algorithms that train a model from known datasets to make predictions, the commonly used MFA algorithm trains a metabolic network with data from isotope tracing experiments and directly outputs the learned information — that is, all fluxes in the network that best fit data3,5 (Fig. 1b). However, as a machine learning algorithm, current MFA methods often lack systematic evaluation and benchmarking, a standard practice in broader machine learning and artificial intelligence applications6. Issues such as algorithmic convergence, flux estimation accuracy and result robustness in MFA studies have been raised but remain largely unaddressed3, limiting the effectiveness and broader adoption of these automated tools in metabolic research.

To advance the capabilities of MFA for complex metabolic networks and extensive isotope tracing datasets, we developed an automated analysis methodology alongside a large-scale metabolic network model. This model comprises over 100 fluxes across key pathways, including glycolysis, the tricarboxylic acid (TCA) cycle, the pentose phosphate pathway (PPP), one-carbon metabolism, and several amino acid (AA) biosynthetic pathways (Fig. 1c, Supplementary Methods). Compared to contemporary MFA tools7,8,9, a notable feature of our methodology is the incorporation of organelle compartmentalization, facilitating accurate quantification of exchange fluxes between mitochondria and cytosol in eukaryotic cells (Fig. 1c). While other tools typically require tens of minutes to obtain a solution7,9, our methodology can generate an optimized solution, with fluxes that accurately explain the labelling pattern from a 13C tracing experiment on cultured cell lines, within 2 s on a desktop computer10 (Supplementary Fig. 1a–d). Nonetheless, a challenge arose from the observation that these optimized solutions could diverge significantly, showing considerable variability in certain net fluxes even with similar loss values (Fig. 1d, Supplementary Fig. 1e,f).To address this problem, we developed an optimization-averaging algorithm that refines the computation process by selecting a subset of solutions with minimal loss (selected solutions) from the pool of optimized solutions and averaging them to produce a new, more stable solution set (averaged solutions) (Fig. 1e, Supplementary Methods). These solutions, along with those generated using the typical strategy used in contemporary software (Supplementary Methods, Supplementary Fig. 1c), were benchmarked using simulated 13C tracing datasets generated from a known flux vector (Supplementary Fig. 2a). The results demonstrated that, relative to the benchmark, the optimization-averaging algorithm effectively reduced flux variability and improved the accuracy of the results in approximating the known flux, even with varying levels of data availability (Fig. 1f, Supplementary Figs. 2b–e and 3a–e).

Abstract Image

利用自动方法从同位素追踪数据中量化代谢活动
代谢通量分析(MFA)是一种基于机器学习原理破译标记模式的计算方法。与典型的机器学习算法从已知数据集训练模型进行预测不同,常用的 MFA 算法利用同位素追踪实验的数据训练代谢网络,并直接输出学习到的信息,即网络中最适合数据的所有通量3,5(图 1b)。然而,作为一种机器学习算法,目前的 MFA 方法往往缺乏系统的评估和基准测试,而这是更广泛的机器学习和人工智能应用中的标准做法6。MFA 研究中的算法收敛性、通量估计准确性和结果稳健性等问题已被提出,但在很大程度上仍未得到解决3,这限制了这些自动化工具在代谢研究中的有效性和广泛应用。该模型包含关键通路中的 100 多个通量,包括糖酵解、三羧酸(TCA)循环、磷酸戊糖通路(PPP)、一碳代谢和几个氨基酸(AA)生物合成通路(图 1c,补充方法)。与当代的 MFA 工具7,8,9 相比,我们的方法的一个显著特点是纳入了细胞器分区,有助于准确量化真核细胞线粒体和细胞膜之间的交换通量(图 1c)。其他工具通常需要几十分钟才能得到一个解决方案7,9,而我们的方法可以在台式电脑上10 在 2 秒钟内生成一个优化解决方案,其通量可以准确解释培养细胞系 13C 追踪实验的标记模式(补充图 1a-d)。为了解决这个问题,我们开发了一种优化-平均算法,该算法通过从优化解决方案池中选择损失最小的解决方案子集(选定解决方案),并对其进行平均,从而生成一个新的、更稳定的解决方案集(平均解决方案)(图 1e,补充方法)来完善计算过程。使用已知通量矢量生成的模拟 13C 追踪数据集(补充图 2a)对这些解决方案以及使用当代软件中使用的典型策略生成的解决方案(补充方法,补充图 1c)进行了基准测试。结果表明,与基准相比,优化平均算法有效地减少了通量的变化,并提高了近似已知通量结果的准确性,即使数据可用性水平不同也是如此(图 1f,补充图 2b-e 和 3a-e)。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Nature metabolism
Nature metabolism ENDOCRINOLOGY & METABOLISM-
CiteScore
27.50
自引率
2.40%
发文量
170
期刊介绍: Nature Metabolism is a peer-reviewed scientific journal that covers a broad range of topics in metabolism research. It aims to advance the understanding of metabolic and homeostatic processes at a cellular and physiological level. The journal publishes research from various fields, including fundamental cell biology, basic biomedical and translational research, and integrative physiology. It focuses on how cellular metabolism affects cellular function, the physiology and homeostasis of organs and tissues, and the regulation of organismal energy homeostasis. It also investigates the molecular pathophysiology of metabolic diseases such as diabetes and obesity, as well as their treatment. Nature Metabolism follows the standards of other Nature-branded journals, with a dedicated team of professional editors, rigorous peer-review process, high standards of copy-editing and production, swift publication, and editorial independence. The journal has a high impact factor, has a certain influence in the international area, and is deeply concerned and cited by the majority of scholars.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信