Machine learning methods for multi-walled carbon nanotubes (MWCNT) genotoxicity prediction

Kotzabasaki, Marianna; Sotiropoulos, Iason; Charitidis, Costas A.; Sarimveis, Haralambos

doi:10.1039/d0na00600a

Cited by 24 publications

(17 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As such, the negative results for the genotoxicity endpoints could be read-across from data-rich to data-poor members MWCNT. An alternative QSAR model for predicting the genotoxicity of the same panel of MWCNT under investigation here and included in the Aschberger study was recently developed by Kotzabasaki et al, [ 44 ]. In contrast to Aschberger et al, [ 43 ] a number of MWCNT from within the overlapping panel were designated genotoxic based on results from in vitro chromosome aberration (micronucleus) assays and the in vivo DNA damage Comet assay.…”

Section: Discussionmentioning

confidence: 99%

“…Several supervised and unsupervised in silico models were applied by the authors to predict this genotoxic outcome based on correlations with a panel of PC properties used as input features. The resulting QSAR model predicted genotoxicity using the 3 input features of mean length, % purity and zeta potential [ 44 ]. Neither grouping outcome presented by these two studies has been validated for biological relevance, as from the MWCNT panel tested only NRCWE006 has been shown, to date, to cause tumour formation in vivo [ 19 ].…”

Section: Discussionmentioning

confidence: 99%

“…high aspect ratio), [ 40 ] and single hazard outcomes (e.g. genotoxicity), [ 43 , 44 ] currently available, expert-led interpretation of grouping and similarity is still required to substantiate group formation in terms of biological relevance and degree of similarity which may differentiate between MWCNT based on critical drivers of long-term human hazard. Recent work by [ 45 ] suggests that in vivo pathological outcomes may be directly linked to the secondary structures and degree of agglomeration of MWCNT which may be predicted from the nominal tube dimensions.…”

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

Grouping MWCNTs based on their similar potential to cause pulmonary hazard after inhalation: a case-study

et al. 2022

View full text Add to dashboard Cite

Background The EU-project GRACIOUS developed an Integrated Approach to Testing and Assessment (IATA) to support grouping high aspect ratio nanomaterials (HARNs) presenting a similar inhalation hazard. Application of grouping reduces the need to assess toxicity on a case-by-case basis and supports read-across of hazard data from substances that have the data required for risk assessment (source) to those that lack such data (target). The HARN IATA, based on the fibre paradigm for pathogenic fibres, facilitates structured data gathering to propose groups of similar HARN and to support read-across by prompting users to address relevant questions regarding HARN morphology, biopersistence and inflammatory potential. The IATA is structured in tiers, allowing grouping decisions to be made using simple in vitro or in silico methods in Tier1 progressing to in vivo approaches at the highest Tier3. Here we present a case-study testing the applicability of GRACIOUS IATA to form an evidence-based group of multiwalled carbon nanotubes (MWCNT) posing a similar predicted fibre-hazard, to support read-across and reduce the burden of toxicity testing. Results The case-study uses data on 15 different MWCNT, obtained from the published literature. By following the IATA, a group of 2 MWCNT was identified (NRCWE006 and NM-401) based on a high degree of similarity. A pairwise similarity assessment was subsequently conducted between the grouped MWCNT to evaluate the potential to conduct read-across and fill data gaps required for regulatory hazard assessment. The similarity assessment, based on expert judgement of Tier 1 assay results, predicts both MWCNT are likely to cause a similar acute in vivo hazard. This result supports the possibility for read-across of sub-chronic and chronic hazard endpoint data for lung fibrosis and carcinogenicity between the 2 grouped MWCNT. The implications of accepting the similarity assessment based on expert judgement of the MWCNT group are considered to stimulate future discussion on the level of similarity between group members considered sufficient to allow regulatory acceptance of a read-across argument. Conclusion This proof-of-concept case-study demonstrates how a grouping hypothesis and IATA may be used to support a nuanced and evidence-based grouping of ‘similar’ MWCNT and the subsequent interpolation of data between group members to streamline the hazard assessment process.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Grouping MWCNTs based on their similar potential to cause pulmonary hazard after inhalation: a case-study

et al. 2022

View full text Add to dashboard Cite

show abstract

“…In case of bioresearch fields, there is also an ethical aspect to consider. Performing experiments on animals is lengthy and costly, while in 2006 Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH) suggests the adoption of alternative research methods, such as in silico [ 40 , 115 ]. In silico methods for nanomaterial design have been emerging with quantitative structure–activity relationship (QSAR) models, providing implications in drug and molecular design of small sequences of organic molecules.…”

Section: Selection Of Nanomaterials Tailored For Improvements In Qual...mentioning

confidence: 99%

Digital Innovation Enabled Nanomaterial Manufacturing; Machine Learning Strategies and Green Perspectives

Konstantopoulos

Koumoulos²,

Charitidis

2022

Nanomaterials

Self Cite

View full text Add to dashboard Cite

Machine learning has been an emerging scientific field serving the modern multidisciplinary needs in the Materials Science and Manufacturing sector. The taxonomy and mapping of nanomaterial properties based on data analytics is going to ensure safe and green manufacturing with consciousness raised on effective resource management. The utilization of predictive modelling tools empowered with artificial intelligence (AI) has proposed novel paths in materials discovery and optimization, while it can further stimulate the cutting-edge and data-driven design of a tailored behavioral profile of nanomaterials to serve the special needs of application environments. The previous knowledge of the physics and mathematical representation of material behaviors, as well as the utilization of already generated testing data, received specific attention by scientists. However, the exploration of available information is not always manageable, and machine intelligence can efficiently (computational resources, time) meet this challenge via high-throughput multidimensional search exploration capabilities. Moreover, the modelling of bio-chemical interactions with the environment and living organisms has been demonstrated to connect chemical structure with acute or tolerable effects upon exposure. Thus, in this review, a summary of recent computational developments is provided with the aim to cover excelling research and present challenges towards unbiased, decentralized, and data-driven decision-making, in relation to increased impact in the field of advanced nanomaterials manufacturing and nanoinformatics, and to indicate the steps required to realize rapid, safe, and circular-by-design nanomaterials.

show abstract

“…To speed up research on CNMs, cutting-edge in silico approaches such as machine-learning [ 57 ] are key enablers to fully exploit large datasets for which innovative methods for efficient management are also crucial [ 58 ]. Another critical step for the wide implementation of CNMs pertains to cost-effective large-scale production [ 59 ], ideally in a sustainable manner, in light of the current climate crisis and the urgent need to lower our impact and preserve the environment.…”

Section: Carbon Nanomaterials (Cnms)mentioning

confidence: 99%

Carbon Graphitization: Towards Greener Alternatives to Develop Nanomaterials for Targeted Drug Delivery

Marin

Marchesan

2022

Biomedicines

View full text Add to dashboard Cite

Carbon nanomaterials have attracted great interest for their unique physico-chemical properties for various applications, including medicine and, in particular, drug delivery, to solve the most challenging unmet clinical needs. Graphitization is a process that has become very popular for their production or modification. However, traditional conditions are energy-demanding; thus, recent efforts have been devoted to the development of greener routes that require lower temperatures or that use waste or byproducts as a carbon source in order to be more sustainable. In this concise review, we analyze the progress made in the last five years in this area, as well as in their development as drug delivery agents, focusing on active targeting, and conclude with a perspective on the future of the field.

show abstract

Machine learning methods for multi-walled carbon nanotubes (MWCNT) genotoxicity prediction

Abstract: Multi-walled carbon nanotubes (MWCNTs) are made of multiple single-walled carbon nanotubes (SWCNTs) which are nested inside one another forming concentric cylinders. These nanomaterials are widely used in industrial and biomedical...

Cited by 24 publications

References 42 publications

Grouping MWCNTs based on their similar potential to cause pulmonary hazard after inhalation: a case-study

Grouping MWCNTs based on their similar potential to cause pulmonary hazard after inhalation: a case-study

Digital Innovation Enabled Nanomaterial Manufacturing; Machine Learning Strategies and Green Perspectives

Carbon Graphitization: Towards Greener Alternatives to Develop Nanomaterials for Targeted Drug Delivery

Contact Info

Product

Resources

About