In silico prediction of chemical acute contact toxicity on honey bees via machine learning methods

Xu, Xuan; Zhao, Piaopiao; Wang, Zhiyuan; Zhang, Xiaoxiao; Wu, Zengrui; Li, Weihua; Tang, Yun; Liu, Guixia

doi:10.1016/j.tiv.2021.105089

Cited by 15 publications

(4 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dominant enzyme-based mechanisms of toxic action in honeybees and their corresponding structural examples are summarized in Table 4. Several important substructures responsible for chemical toxicity in honeybee were identified in the study by Xu et al 239 and the study by Li et al 240 Representative reported structural alerts are summarized in Fig. 27, which consist of five categories, as follows: (i) phosphoric acid derivatives and phosphanamine, which potentially exhibit inhibition activity towards AChE or cholinergic receptors, (2) core structure of pyrethroids, which exert bee toxicity by acting on sodium channel, (3) allyl chloride or mono-chloroalkene, which are also genotoxic fragments, (4) nitramide groups frequently occurring in neonicotinoids, which that can favor the inhibition of nAChR, and (5) other substructures, such as 1-(2-isopropoxyphenoxy)-N-methylmethanamine and (E)-acetaldehyde O-methylcarbamoyl oxime.…”

Section: Structural Features Associated With Specific Ecotoxicitymentioning

confidence: 99%

Toxicological data bank bridges the gap between environmental risk assessment and green organic chemical design in One Health world

et al. 2023

View full text Add to dashboard Cite

show abstract

Section: Structural Features Associated With Specific Ecotoxicitymentioning

confidence: 99%

Toxicological data bank bridges the gap between environmental risk assessment and green organic chemical design in One Health world

et al. 2023

View full text Add to dashboard Cite

show abstract

“…An interesting point with the latest development in modelling approach remains in the study of sub-lethal doses of pesticides and their combined effect on the colonies [ 270 ] as well as the long-term effect of antibiotics on bees [ 271 ]. Even acute contact toxicity was predicted recently [ 272 ], advocating for a deeper connection between in silico, in vitro, semi-field and field experiments. In the end, the host-parasite relationship should be part of an integrative view and inserted in a wider picture where other variables like poor nutrition, crop pesticides or pathogens are assessed to identify if they act in synergism or antagonism inside the hive [ 273 ].…”

Section: From the Laboratory To The Field: How To Be Realistic?mentioning

confidence: 99%

Varroa destructor from the Laboratory to the Field: Control, Biocontrol and IPM Perspectives—A Review

Vilarem

Piou

Vogelweith³

et al. 2021

Insects

View full text Add to dashboard Cite

Varroadestructor is a real challenger for beekeepers and scientists: fragile out of the hive, tenacious inside a bee colony. From all the research done on the topic, we have learned that a better understanding of this organism in its relationship with the bee but also for itself is necessary. Its biology relies mostly on semiochemicals for reproduction, nutrition, or orientation. Many treatments have been developed over the years based on hard or soft acaricides or even on biocontrol techniques. To date, no real sustainable solution exists to reduce the pressure of the mite without creating resistances or harming honeybees. Consequently, the development of alternative disruptive tools against the parasitic life cycle remains open. It requires the combination of both laboratory and field results through a holistic approach based on health biomarkers. Here, we advocate for a more integrative vision of V. destructor research, where in vitro and field studies are more systematically compared and compiled. Therefore, after a brief state-of-the-art about the mite’s life cycle, we discuss what has been done and what can be done from the laboratory to the field against V. destructor through an integrative approach.

show abstract

“…Generally, training machine learning models to predict the toxicity of compounds to biological organisms is an active area of research [43,44]. And, indeed, open data from bee toxicity experiments [45][46][47][48][49][50][51][52][53][54][55][56][57] have been leveraged to train machine learning models to computationally predict the toxicity of pesticides to bees [58][59][60][61][62][63][64].…”

Section: Introduction 1pesticide Toxicity To Beesmentioning

confidence: 99%

Classifying the toxicity of pesticides to honey bees via support vector machines with random walk graph kernels

Yang

Henle

Fern

et al. 2022

The Journal of Chemical Physics

View full text Add to dashboard Cite

Pesticides benefit agriculture by increasing crop yield, quality, and security. However, pesticides may inadvertently harm bees, which are valuable as pollinators. Thus, candidate pesticides in development pipelines must be assessed for toxicity to bees. Leveraging a dataset of 382 molecules with toxicity labels from honey bee exposure experiments, we train a support vector machine (SVM) to predict the toxicity of pesticides to honey bees. We compare two representations of the pesticide molecules: (i) a random walk feature vector listing counts of length- L walks on the molecular graph with each vertex- and edge-label sequence and (ii) the Molecular ACCess System (MACCS) structural key fingerprint (FP), a bit vector indicating the presence/absence of a list of pre-defined subgraph patterns in the molecular graph. We explicitly construct the MACCS FPs but rely on the fixed-length- L random walk graph kernel (RWGK) in place of the dot product for the random walk representation. The L-RWGK-SVM achieves an accuracy, precision, recall, and F1 score (mean over 2000 runs) of 0.81, 0.68, 0.71, and 0.69, respectively, on the test data set—with L = 4 being the mode optimal walk length. The MACCS-FP-SVM performs on par/marginally better than the L-RWGK-SVM, lends more interpretability, but varies more in performance. We interpret the MACCS-FP-SVM by illuminating which subgraph patterns in the molecules tend to strongly push them toward the toxic/non-toxic side of the separating hyperplane.

show abstract

In silico prediction of chemical acute contact toxicity on honey bees via machine learning methods

Cited by 15 publications

References 39 publications

Toxicological data bank bridges the gap between environmental risk assessment and green organic chemical design in One Health world

Toxicological data bank bridges the gap between environmental risk assessment and green organic chemical design in One Health world

Varroa destructor from the Laboratory to the Field: Control, Biocontrol and IPM Perspectives—A Review

Classifying the toxicity of pesticides to honey bees via support vector machines with random walk graph kernels

Contact Info

Product

Resources

About