Reusability report: Evaluating reproducibility and reusability of a fine-tuned model to predict drug response in cancer patient samples

So, Emily; Yu, Fengqing; Wang, Bo; Haibe‐Kains, Benjamin

doi:10.1038/s42256-023-00688-4

Cited by 4 publications

(2 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To obtain modeling data with clear functional relationships, ve variables related to the x are dened, see eqn (1)- (5). With the tolerance of 2, the arithmetic sequence in the range of (400, 1000), (488, 888), (20,400), and (1000, 1400) was generated as the independent variable data for the training set, the test (I) set, the test (B) set, and the test (F) set, respectively.…”

Section: Mathematical Model Extrapolation Testmentioning

confidence: 99%

“…Machine learning (ML) has made impressive achievements in substance discovery, data analysis, and image processing over the past decades, accelerating advances in elds as numerous as earth & life science, [1][2][3] communications & transportation, [4][5][6][7][8][9][10] and chemistry & medicine. [11][12][13][14][15][16][17][18][19][20] As a spotlight to the eld of chemistry, ML provides experimentalists with advice on selecting target molecules for synthesis by predicting physicochemical properties, [11][12][13][14][15] biological effects, [16][17][18] and reaction routes.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Extrapolation validation (EV): a universal validation method for mitigating machine learning extrapolation risk

Yu,

Zhou,

Wang

et al. 2024

Digital Discovery

View full text Add to dashboard Cite

show abstract

Section: Mathematical Model Extrapolation Testmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Extrapolation validation (EV): a universal validation method for mitigating machine learning extrapolation risk

Yu,

Zhou,

Wang

et al. 2024

Digital Discovery

View full text Add to dashboard Cite

show abstract

Artificial intelligence in food and nutrition evidence: The challenges and opportunities

Bailey,

MacFarlane,

Field

et al. 2024

PNAS Nexus

View full text Add to dashboard Cite

Science-informed decisions are best guided by the objective synthesis of the totality of evidence around a particular question and assessing its trustworthiness through systematic processes. However, there are major barriers and challenges that limit science-informed food and nutrition policy, practice, and guidance. First, insufficient evidence, primarily due to acquisition cost of generating high-quality data, and the complexity of the diet-disease relationship. Furthermore, the sheer number of systematic reviews needed across the entire agriculture and food value chain, and the cost and time required to conduct them, can delay the translation of science to policy. Artificial intelligence (AI) offers the opportunity: 1) to better understand the complex etiology of diet-related chronic diseases; 2) to bring more precision to our understanding of the variation among individuals in the diet-chronic disease relationship; 3) to provide new types of computed data related to the efficacy and effectiveness of nutrition/food interventions in health promotion; and 4) to automate the generation of systematic reviews that support timely decisions. These advances include the acquisition and synthesis of heterogeneous and multimodal datasets. This perspective summarizes findings from a meeting of experts convened at the National Academy of Sciences, Engineering and Medicine. The purpose of the meeting was to examine the current state and future potential of AI in generating new types of computed data as well as automating the generation of systematic reviews to support evidence-based food and nutrition policy, practice and guidance.

show abstract

A Universal Validation Method for Mitigating Machine Learning Extrapolation Risk

Yan,

Yu,

Zhou

et al. 2023

Preprint

View full text Add to dashboard Cite

Machine Learning (ML) can provide decision-making advice for major challenges in science and engineering, and its rapid development has led to advances in fields like chemistry & medicine, earth & life, and communications & transportation. Grasping the trustworthiness of the decision-making advice given by ML models remains challenging, especially when applying them to samples outside the domain-of-application. Here, an untrustworthy application situation (i.e., complete extrapolation-failure) that would occur in models developed by ML methods involving tree algorithms is confirmed, and the root cause is revealed. Further, a universal extrapolation risk evaluation scheme, termed the extrapolation validation (EV) method, is proposed, which is not restricted to specific ML methods and model architecture in its applicability. The EV method quantitatively evaluates the extrapolation ability of 11 popularly applied ML methods and digitalizes the extrapolation risk arising from variations of the independent variables in each method. Meanwhile, the EV method provides insights and solutions for evaluating the reliability of out-of-distribution sample prediction and selecting trustworthy ML methods.

show abstract

Reusability report: Evaluating reproducibility and reusability of a fine-tuned model to predict drug response in cancer patient samples

Cited by 4 publications

References 20 publications

Extrapolation validation (EV): a universal validation method for mitigating machine learning extrapolation risk

Extrapolation validation (EV): a universal validation method for mitigating machine learning extrapolation risk

Artificial intelligence in food and nutrition evidence: The challenges and opportunities

A Universal Validation Method for Mitigating Machine Learning Extrapolation Risk

Contact Info

Product

Resources

About