Integrative In Silico Toxicity Assessment of Chlorfenapyr Using AI-Driven Platforms

Ünzile Yaman

Makale detayları

Integrative In Silico Toxicity Assessment of Chlorfenapyr Using AI-Driven Platforms

Yazarlar:

Ünzile Yaman

Yayıncı:

Izmir Katip Celebi University

Ana Başlıklar:

Modelleme ve Simülasyon | Yapay Zeka (Diğer)

Dergi:

Journal of Artificial Intelligence and Data Science Vol.5 Issue 1, pp.44-52

Belge Tipi:

Makale

Yayın Tarihi:

6/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001351932

pISSN:

2791-8335

Anahtar kelimeler:

Özet:

 Chlorfenapyr is a pyrrole-class pesticide with a unique mechanism of action that disrupts mitochondrial oxidative phosphorylation. Despite its broad-spectrum insecticidal use, publicly available toxicological data on chlorfenapyr remain limited, particularly regarding organ-specific and long-term effects. To address this data gap, the present study implements a multi-model in silico toxicity assessment using three AI-based platformsSwissADME, ProTox-II, and ADMETlab 2.0to predict key toxicokinetic and toxicodynamic properties from the compounds SMILES representation. Physicochemical and pharmacokinetic parameters such as molecular weight, lipophilicity, gastrointestinal absorption, and cytochrome P450 inhibition were consistently predicted across platforms. However, notable discrepancies emerged in bloodbrain barrier BBB permeability and hepatotoxicity outcomes. Acute toxicity was estimated with a predicted LD₅₀ of 55 mgkg Class 3, while organ-specific risks included neurotoxicity and respiratory toxicity. Both platforms highlighted mitochondrial membrane potential disruption and oxidative stress pathways as probable mechanisms of toxicity. Toxicophore analysis further revealed substructures associated with non-genotoxic carcinogenicity, aquatic toxicity, and poor biodegradability, raising environmental safety concerns. By combining complementary model outputs, this AI-supported approach allows for scalable, reproducible, and ethically favorable screening of chemical hazards. The findings demonstrate that multi-endpoint in silico toxicology workflows can effectively identify early warning signals of compound toxicity and guide future experimental prioritiesparticularly for chemicals like chlorfenapyr, where experimental data are scarce and regulatory insight is urgently needed.

 Chlorfenapyr is a pyrrole-class pesticide with a unique mechanism of action that disrupts mitochondrial oxidative phosphorylation. Despite its broad-spectrum insecticidal use, publicly available toxicological data on chlorfenapyr remain limited, particularly regarding organ-specific and long-term effects. To address this data gap, the present study implements a multi-model in silico toxicity assessment using three AI-based platformsSwissADME, ProTox-II, and ADMETlab 2.0to predict key toxicokinetic and toxicodynamic properties from the compounds SMILES representation. Physicochemical and pharmacokinetic parameters such as molecular weight, lipophilicity, gastrointestinal absorption, and cytochrome P450 inhibition were consistently predicted across platforms. However, notable discrepancies emerged in bloodbrain barrier BBB permeability and hepatotoxicity outcomes. Acute toxicity was estimated with a predicted LD₅₀ of 55 mgkg Class 3, while organ-specific risks included neurotoxicity and respiratory toxicity. Both platforms highlighted mitochondrial membrane potential disruption and oxidative stress pathways as probable mechanisms of toxicity. Toxicophore analysis further revealed substructures associated with non-genotoxic carcinogenicity, aquatic toxicity, and poor biodegradability, raising environmental safety concerns. By combining complementary model outputs, this AI-supported approach allows for scalable, reproducible, and ethically favorable screening of chemical hazards. The findings demonstrate that multi-endpoint in silico toxicology workflows can effectively identify early warning signals of compound toxicity and guide future experimental prioritiesparticularly for chemicals like chlorfenapyr, where experimental data are scarce and regulatory insight is urgently needed.

 Chlorfenapyr is a pyrrole-class pesticide with a unique mechanism of action that disrupts mitochondrial oxidative phosphorylation. Despite its broad-spectrum insecticidal use, publicly available toxicological data on chlorfenapyr remain limited, particularly regarding organ-specific and long-term effects. To address this data gap, the present study implements a multi-model in silico toxicity assessment using three AI-based platformsSwissADME, ProTox-II, and ADMETlab 2.0to predict key toxicokinetic and toxicodynamic properties from the compounds SMILES representation. Physicochemical and pharmacokinetic parameters such as molecular weight, lipophilicity, gastrointestinal absorption, and cytochrome P450 inhibition were consistently predicted across platforms. However, notable discrepancies emerged in bloodbrain barrier BBB permeability and hepatotoxicity outcomes. Acute toxicity was estimated with a predicted LD₅₀ of 55 mgkg Class 3, while organ-specific risks included neurotoxicity and respiratory toxicity. Both platforms highlighted mitochondrial membrane potential disruption and oxidative stress pathways as probable mechanisms of toxicity. Toxicophore analysis further revealed substructures associated with non-genotoxic carcinogenicity, aquatic toxicity, and poor biodegradability, raising environmental safety concerns. By combining complementary model outputs, this AI-supported approach allows for scalable, reproducible, and ethically favorable screening of chemical hazards. The findings demonstrate that multi-endpoint in silico toxicology workflows can effectively identify early warning signals of compound toxicity and guide future experimental prioritiesparticularly for chemicals like chlorfenapyr, where experimental data are scarce and regulatory insight is urgently needed.

 Chlorfenapyr, mitokondriyal oksidatif fosforilasyonu bozan özgün etki mekanizmasına sahip pirrol sınıfı bir pestisittir. Geniş spektrumlu insektisit olarak yaygın kullanımına rağmen, chlorfenapyrin toksikolojik profiline dair veriler sınırlıdır; özellikle organ spesifik etkiler ve uzun vadeli toksisite açısından önemli bilgi boşlukları bulunmaktadır. Bu boşluğu doldurmak amacıyla, bu çalışma chlorfenapyr bileşiğinin SMILES koduna dayalı olarak SwissADME, ProTox-II ve ADMETlab 2.0 gibi üç yapay zekâ destekli platform kullanılarak çok modeli bir in silico toksisite değerlendirmesi gerçekleştirmiştir. Moleküler ağırlık, lipofiliklik, gastrointestinal emilim ve sitokrom P450 inhibisyonu gibi fizikokimyasal ve farmakokinetik parametreler, platformlar arasında tutarlı biçimde öngörülmüştür. Ancak, kan-beyin bariyeri geçirgenliği ve hepatotoksisite tahminlerinde belirgin tutarsızlıklar gözlemlenmiştir. Akut toksisite, tahmini LD₅₀ değeri 55 mgkg Sınıf 3 olarak değerlendirilmiştir; organ düzeyinde toksik etkiler arasında ise nörotoksisite ve solunum toksisitesi öne çıkmaktadır. Her iki platform da mitokondriyal membran potansiyeli bozulması ve oksidatif stres yollarının olası toksisite mekanizmaları olduğunu vurgulamıştır. Toksikofor analizi, genotoksik olmayan karsinojenite, sucul toksisite ve biyobozunmazlık ile ilişkili alt yapılar ortaya koyarak çevresel güvenlik açısından da endişelere işaret etmektedir. Tamamlayıcı modellerin çıktılarının birleştirilmesi, bu yapay zekâ destekli yaklaşımı kimyasal tehlike taramalarında ölçeklenebilir, tekrarlanabilir ve etik açıdan uygun hale getirmektedir. Bulgular, çok uçlu in silico toksikoloji iş akışlarının, bileşik toksisitesine dair erken uyarı sinyallerini etkin biçimde saptayabileceğini ve özellikle chlorfenapyr gibi deneysel verisi kısıtlı kimyasallar için gelecekteki araştırma önceliklerini belirlemede yol gösterici olabileceğini göstermektedir.

Makale detayları

Integrative In Silico Toxicity Assessment of Chlorfenapyr Using AI-Driven Platforms

Yazarlar:

Ünzile Yaman

Yayıncı:

Izmir Katip Celebi University

Ana Başlıklar:

Modelleme ve Simülasyon | Yapay Zeka (Diğer)

Dergi:

Journal of Artificial Intelligence and Data Science Vol.5 Issue 1, pp.44-52

Belge Tipi:

Makale

Yayın Tarihi:

6/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001351932

pISSN:

2791-8335

ADMETlab 2.0

Chlorfenapyr

Computational Toxicology

ProTox-II

Pesticide Risk Assessment

SwissADMEADMETlab2.0

Chlorfenapyr

hesaplamalı toksikoloji

ProTox-II

pestisit risk değerlendirmesi

SwissADME

Sayı

Makale detayları

Yazarlar: Ünzile Yaman

Yayıncı: Izmir Katip Celebi University

Ana Başlıklar: Modelleme ve Simülasyon | Yapay Zeka (Diğer)

Dergi: Journal of Artificial Intelligence and Data Science Vol.5 Issue 1, pp.44-52

Belge Tipi: Makale

Yayın Tarihi: 6/2025

Belge Dili: İngilizce

Document ID (Org No.): 2001351932

pISSN: 2791-8335

Sayı

Yazarlar:

Ünzile Yaman

Yayıncı:

Izmir Katip Celebi University

Ana Başlıklar:

Modelleme ve Simülasyon | Yapay Zeka (Diğer)

Dergi:

Journal of Artificial Intelligence and Data Science Vol.5 Issue 1, pp.44-52

Belge Tipi:

Makale

Yayın Tarihi:

6/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001351932

pISSN:

2791-8335