Recent advances in delivery systems optimization using machine learning approaches

Yakoubi, Sana; Kobayashi, Isao; Uemura, Kunihiko; Nakajima, Mitsutoshi; Hiroko, Isoda; Neves, Marcos A.

doi:10.1016/j.cep.2023.109352

Cited by 15 publications

(8 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given that the decision-tree based models most effectively analyzed the dataset, it suggests that the relationships between formulation parameters and transfection efficiency are non-linear and complex otherwise regression models would have performed equally well. Similar results have been reported in other comparable tasks where decision-tree based models outperform various other model architectures [35][36][37][38][39][40] . We achieved minimal error (MAE < 0.11), strong Pearson's correlation (r > 0.8), and moderate to strong Spearman's rank correlation (r > 0.6) between the predictions and hold-out experimental values across all cell types (Supplementary Table 4, Supplementary Figure 4).…”

Section: Decision Tree-based Algorithms Describe Complex Non-linear R...supporting

confidence: 87%

Machine Learning Elucidates Design Features of Plasmid DNA Lipid Nanoparticles for Cell Type-Preferential Transfection

Cheng,

Zhu,

et al. 2023

Preprint

View full text Add to dashboard Cite

For cell and gene therapies to become more broadly accessible, it is critical to develop and optimize non-viral cell type-preferential gene carriers such as lipid nanoparticles (LNPs).Despite the effectiveness of high throughput screening (HTS) approaches in expediting LNP discovery, they are often costly, labor-intensive, and often do not provide actionable LNP design rules that focus screening efforts on the most relevant chemical and formulation parameters. Here we employed a machine learning (ML) workflow using well-curated plasmid DNA LNP transfection datasets across six cell types to maximize chemical insights from HTS studies and has achieved predictions with 5–9% error on average depending on cell type. By applying Shapley additive explanations to our ML models, we unveiled composition-function relationships dictating cell type-preferential LNP transfection efficiency. Notably, we identified consistent LNP composition parameters that enhancein vitrotransfection efficiency across diverse cell types, such as ionizable to helper lipid ratios near 1:1 or 10:1 and the incorporation of cationic/zwitterionic helper lipids. In addition, several parameters were found to modulate cell type-preferentiality, including the ionizable and helper lipid total molar percentage, N/P ratio, cholesterol to PEGylated lipid ratio, and the chemical identity of the helper lipid. This study leverages HTS of compositionally diverse LNP libraries and ML analysis to understand the interactions between lipid components in LNP formulations; and offers fundamental insights that contribute to the establishment of unique sets of LNP compositions tailored for cell type-preferential transfection.

show abstract

Section: Decision Tree-based Algorithms Describe Complex Non-linear R...supporting

confidence: 87%

Machine Learning Elucidates Design Features of Plasmid DNA Lipid Nanoparticles for Cell Type-Preferential Transfection

Cheng,

Zhu,

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Researchers are investigating the complexities of machine learning approaches such as artificial neural networks (ANNs), genetic algorithms (GA), support vector machines (SVM), random forest, K-mean clustering, and decision tress to determine how they might be useful in designing improved food-delivery systems [136,137]. They are focusing on developing a ready-made machine-learning approach which can optimize all the parameters for nanoemulsion formulation.…”

Section: Application Of Artificial Intelligence (Ai) and Machine Lear...mentioning

confidence: 99%

“…ANNs can be primed with all the mathematical algorithms and can be trained to detect patterns in data, which can be utilized for prediction and decisionmaking. The power of ANNs has been exploited in the formulation of nanoemulsion and its optimization to predict the size of droplets in nanoemulsions, their stability, and various other parameters of the final emulsion depending upon the input variables such as type of essential oil (or bioactive compound), type of surfactant/co-surfactant used, and processing conditions [137,138]. A research study was conducted to optimize and anticipate the output of Curcuma longa L. essential oil in various agro-climatic zones through the use of ANNs.…”

Section: Application Of Artificial Intelligence (Ai) and Machine Lear...mentioning

confidence: 99%

“…The selected formulations would be allowed to reproduce using key variables of GA, mutation (in which parent formulation is altered by sudden change in one or more components), and crossovers (in which two formulations are combined to reproduce a new formulation). The success rate of new offspring (formulations) will again be evaluated using the fitness score, and this process shall continue until end point criteria are fulfilled (stop signal may be defined as a set of generations or a stable formulation) [137]. W/O nanoemulsion was fabricated by combining UC (ultrasonic cavitation) and ISD (isothermal dilution process).…”

Section: Application Of Artificial Intelligence (Ai) and Machine Lear...mentioning

confidence: 99%

See 1 more Smart Citation

Advances in Designing Essential Oil Nanoformulations: An Integrative Approach to Mathematical Modeling with Potential Application in Food Preservation

Soni,

Yadav,

Maurya

et al. 2023

Foods

View full text Add to dashboard Cite

Preservation of foods, along with health and safety issues, is a growing concern in the current generation. Essential oils have emerged as a natural means for the long-term protection of foods along with the maintenance of their qualities. Direct applications of essential oils have posed various constraints to the food system and also have limitations in application; hence, encapsulation of essential oils into biopolymers has been recognized as a cutting-edge technology to overcome these challenges. This article presents and evaluates the strategies for the development of encapsulated essential oils on the basis of fascination with the modeling and shuffling of various biopolymers, surfactants, and co-surfactants, along with the utilization of different fabrication processes. Artificial intelligence and machine learning have enabled the preparation of different nanoemulsion formulations, synthesis strategies, stability, and release kinetics of essential oils or their bioactive components from nanoemulsions with improved efficacy in food systems. Different mathematical models for the stability and delivery kinetics of essential oils in food systems have also been discussed. The article also explains the advanced application of modeling-based encapsulation strategies on the preservation of a variety of food commodities with their intended implication in food and agricultural industries.

show abstract

“…Genetic algorithms have both global and local search capabilities because they are implemented through constructed fitness functions and genetic operations [6][7][8] . However, although genetic algorithms can achieve balanced search and perform well in solving many complex problems, they often suffer from premature or non-converging problems.…”

Section: ．Problems With Genetic Algorithmsmentioning

confidence: 99%

An improved genetic algorithm based on quantitative data-network vulnerability testing as a sample

Li,

Yan,

Fan

et al. 2023

Fourth International Conference on Signal Processing and Computer Science (SPCS 2023)

View full text Add to dashboard Cite

Recent advances in delivery systems optimization using machine learning approaches

Cited by 15 publications

References 42 publications

Machine Learning Elucidates Design Features of Plasmid DNA Lipid Nanoparticles for Cell Type-Preferential Transfection

Machine Learning Elucidates Design Features of Plasmid DNA Lipid Nanoparticles for Cell Type-Preferential Transfection

Advances in Designing Essential Oil Nanoformulations: An Integrative Approach to Mathematical Modeling with Potential Application in Food Preservation

An improved genetic algorithm based on quantitative data-network vulnerability testing as a sample

Contact Info

Product

Resources

About