Impurity Profiling - Revision history

Kirilas: /* ICH M7 classification algorithm */

2025-06-26T07:49:53Z

ICH M7 classification algorithm

← Older revision		Revision as of 07:49, 26 June 2025
Line 86:		Line 86:

	In accordance with this statement, the classification algorithm involves both statistical and rule-based approaches:		In accordance with this statement, the classification algorithm involves both statistical and rule-based approaches:
	# Primary statistical model – this is the Ames test model ("Bacterial composite" mutagenicity endpoint), which is supported by a QMRF document prepared in accordance with QSAR Assessment Framework (QAF) guidelines [https://perceptahelp.acdlabs.com/help_v2025/~~docs~~/~~ACD_Ames_Test_QMRF~~.pdf]. This model predicts the probability of positive Ames test outcome and Reliability Index (RI) of prediction. This output is always provided in ICH M7 Classification PDF reports.		# Primary statistical model – this is the Ames test model ("Bacterial composite" mutagenicity endpoint), which is supported by a QMRF document prepared in accordance with QSAR Assessment Framework (QAF) guidelines [https://perceptahelp.acdlabs.com/help_v2025/pdf/QMRF_Ames_Test.pdf]. This model predicts the probability of positive Ames test outcome and Reliability Index (RI) of prediction. This output is always provided in ICH M7 Classification PDF reports.
	# A set of auxiliary statistical models – those predict probabilities of hazardous outcomes (with RI) in a variety of other genotoxicity/carcinogenicity related endpoints. The full list is available [[Impurity_Profiling#Experimental_Data\|below]]. These are '''never used''' to prove safety of tested compounds. On the contrary, they are only used to express '''extra concern''', in cases when they produce confident positive predictions.		# A set of auxiliary statistical models – those predict probabilities of hazardous outcomes (with RI) in a variety of other genotoxicity/carcinogenicity related endpoints. The full list is available [[Impurity_Profiling#Experimental_Data\|below]]. These are '''never used''' to prove safety of tested compounds. On the contrary, they are only used to express '''extra concern''', in cases when they produce confident positive predictions.
	# A rule-based system that identifies alerting groups potentially responsible for genotoxic effect and highlight them on the molecular structures.		# A rule-based system that identifies alerting groups potentially responsible for genotoxic effect and highlight them on the molecular structures.
Line 100:		Line 100:
	* Other mitigating factors		* Other mitigating factors

	The algorithm summarizes the findings into an automatic assignment of ICH M7 Class 3 or 5, ~~but~~ all collected information is presented in the PDF report for human review of the results.		The algorithm summarizes the findings into an automatic assignment of ICH M7 Class 3 or 5.

			Based on the considerations above, even if statistical prediction is formally outside AD, the raw numbers may still indicate low probability of mutagenicity. If there are also no structural hazards and no concerns about similar compounds, so that nothing points to any kind of potential safety issues – this may still result in Class 5 prediction.

			In any case, all collected information is presented in the PDF report for human review of the results.

	=== Experimental Data ===		=== Experimental Data ===

Kirilas: Updated for 2025 release

2025-06-18T11:45:54Z

Updated for 2025 release

← Older revision		Revision as of 11:45, 18 June 2025
Line 71:		Line 71:
	\|}		\|}

	The classification algorithm has been developed with the intent to mimic the logic of human expert evaluation. Each classification output is supplemented with reasoning that had led to assignment of a particular class and recommendations regarding appropriate control measures for that class of impurities. When classification cannot be made on the ~~basis~~ of ~~available~~ experimental data ~~alone~~, ~~further evaluation~~ is ~~performed using~~ WOE ~~(weight of evidence~~) approach ~~involving~~:		The classification algorithm has been developed with the intent to mimic the logic of human expert evaluation. Each classification output is supplemented with reasoning that had led to assignment of a particular class and recommendations regarding appropriate control measures for that class of impurities.
	* ~~The~~ probability of hazardous effects reported by statistical models and ~~confidence~~ of predictions
			'''Notes:'''
			# Assignment of ICH M7 Class 4 requires establishing the link between parent and derivative compounds, and therefore, cannot be performed when viewing compounds one by one in the expert module interface. Full-featured classification including Class 4 can be performed in the following ways:
			#* Calculate "ICH M7 Classification" in Spreadsheet workspace and indicate the ID of the parent compound in the active project as the required parameter (set ID = 0 to consider the compounds one by one without account for potential parent-derivative relationships)
			#* Generate Batch PDF reports for all compounds in the project and select the option to use the first visible record as the parent molecule
			# When a definitive classification cannot be made, ICH M7 Class is reported as Inconclusive (rendered as 0 in Spreadsheet workspace) – this should be treated similarly to Class 3 compounds, as requiring further attention.

			===ICH M7 classification algorithm===

			ICH M7 guidelines specifically state that:

			''Two (Q)SAR prediction methodologies that complement each other should be applied. One methodology should be expert rule-based and the second methodology should be statistical-based.''

			In accordance with this statement, the classification algorithm involves both statistical and rule-based approaches:
			# Primary statistical model – this is the Ames test model ("Bacterial composite" mutagenicity endpoint), which is supported by a QMRF document prepared in accordance with QSAR Assessment Framework (QAF) guidelines [https://perceptahelp.acdlabs.com/help_v2025/docs/ACD_Ames_Test_QMRF.pdf]. This model predicts the probability of positive Ames test outcome and Reliability Index (RI) of prediction. This output is always provided in ICH M7 Classification PDF reports.
			# A set of auxiliary statistical models – those predict probabilities of hazardous outcomes (with RI) in a variety of other genotoxicity/carcinogenicity related endpoints. The full list is available [[Impurity_Profiling#Experimental_Data\|below]]. These are '''never used''' to prove safety of tested compounds. On the contrary, they are only used to express '''extra concern''', in cases when they produce confident positive predictions.
			# A rule-based system that identifies alerting groups potentially responsible for genotoxic effect and highlight them on the molecular structures.

			The algorithm first checks availability of experimental data for the tested compound in the built-in databases to determine whether it can be assigned to ICH M7 Class 1 or 2.

			Then, the algorithm employs a very conservative approach to determine whether the compound can be confidently classified as Ames Positive or Negative on the basis of calculated probability and RI. This is summarized in the ICH M7 report as the "Call" value. Without going into technical details, the compound must have both very low predicted probability and sufficiently high RI to obtain a confident Negative call (Positive is analogous), otherwise it is reported as Undefined, even if it is covered by the Applicability Domain.

			Finally, if there are alerting groups, or ambiguous predictions, the algorithm employs Weight Of Evidence (WOE) approach that involves all available information, such as:
			* Raw predicted values - the probability of hazardous effects reported by statistical models and reliability of predictions
	* Presence of alerting groups known from the literature		* Presence of alerting groups known from the literature
	* Evidence from experimental data for the most similar compounds from the built-in database		* Evidence from experimental data for the most similar compounds from the built-in database
	* Other mitigating factors		* Other mitigating factors

	~~'''''Note:''''' Currently, Percepta can only assign Class 4 when~~ ICH M7 ~~Classification~~ is ~~calculated~~ in ~~Spreadsheet workspace with indicated ID~~ of the parent compound in the active project. Calculations in the Expert module UI (and in Spreadsheet with provided ID = 0) consider the compounds one by one without account for potential parent-derivative relationships. Also, when a definitive classification cannot be made, ICH M7 Class is reported as Inconclusive (rendered as 0 in Spreadsheet workspace) – this should be treated similarly to Class 3 compounds, as requiring further attention.		The algorithm summarizes the findings into an automatic assignment of ICH M7 Class 3 or 5, but all collected information is presented in the PDF report for human review of the results.

	=== Experimental Data ===		=== Experimental Data ===

Kirilas: Updated for 2023 release

2023-07-26T10:53:01Z

Updated for 2023 release

← Older revision		Revision as of 10:53, 26 July 2023
Line 77:		Line 77:
	* Other mitigating factors		* Other mitigating factors

	'''''Note:''''' Currently, ~~ACD/~~Percepta only ~~considers~~ the compounds one by one without account for potential parent-derivative relationships~~, and therefore, does not perform Class 4 assignment~~. Also, when a definitive classification cannot be made, ICH M7 Class is reported as Inconclusive (rendered as 0 in Spreadsheet workspace) – this should be treated similarly to Class 3 compounds, as requiring further attention.		'''''Note:''''' Currently, Percepta can only assign Class 4 when ICH M7 Classification is calculated in Spreadsheet workspace with indicated ID of the parent compound in the active project. Calculations in the Expert module UI (and in Spreadsheet with provided ID = 0) consider the compounds one by one without account for potential parent-derivative relationships. Also, when a definitive classification cannot be made, ICH M7 Class is reported as Inconclusive (rendered as 0 in Spreadsheet workspace) – this should be treated similarly to Class 3 compounds, as requiring further attention.

	=== Experimental Data ===		=== Experimental Data ===

Kirilas: Reformatted references to active links

2022-10-20T09:19:01Z

Reformatted references to active links

← Older revision		Revision as of 09:19, 20 October 2022
Line 83:		Line 83:
	A complete list of modeled endpoints is provided in the Table, while the data sources are briefly described below.<br />		A complete list of modeled endpoints is provided in the Table, while the data sources are briefly described below.<br />

	'''Genetic toxicity''': data sets for standard assays reflecting different mechanisms of genetic damage were obtained from the FDA. Gene mutation tests and techniques detecting clastogenic/aneugenic effects are included. Data was collected from EPA GENE-TOX database and scientific literature ~~[1]~~.		'''Genetic toxicity''': data sets for standard assays reflecting different mechanisms of genetic damage were obtained from the FDA. Gene mutation tests and techniques detecting clastogenic/aneugenic effects are included. Data was collected from EPA GENE-TOX database [https://www.nlm.nih.gov/databases/download/genetox.html] and scientific literature.

	'''Carcinogenicity''': results of chronic (two-year term) carcinogenicity studies in rats and mice were received from FDA. This data was based on NTP technical reports, IARC monographs, Carcinogenic Potency DataBase [2] and other publicly available sources. Raw data was converted to binary classification using a weight of evidence (WOE) approach [1]. Classification using the WOE threshold corresponding to “potent carcinogens” was used to build the models in the current study.		'''Carcinogenicity''': results of chronic (two-year term) carcinogenicity studies in rats and mice were received from FDA. This data was based on NTP technical reports, IARC monographs, Carcinogenic Potency DataBase [https://pubmed.ncbi.nlm.nih.gov/15800034] and other publicly available sources. Raw data was converted to binary classification using a weight of evidence (WOE) approach [https://pubmed.ncbi.nlm.nih.gov/17207562]. Classification using the WOE threshold corresponding to “potent carcinogens” was used to build the models in the current study.

	'''Reproductive toxicity''': experimental data characterizing the potential for endocrine system disruption due to Estrogen receptor α binding were acquired from ChEMBL database [3]. Compounds were classified as binders/non-binders on the basis of their relative binding affinities (RBA) compared to reference ligand estradiol. Two cut-offs were used: LogRBA > -3 (“general binding”), and LogRBA > 0 (“strong binding”)<br />		'''Reproductive toxicity''': experimental data characterizing the potential for endocrine system disruption due to Estrogen receptor α binding were acquired from ChEMBL database [https://www.ebi.ac.uk/chembl] (Target ID 206). Compounds were classified as binders/non-binders on the basis of their relative binding affinities (RBA) compared to reference ligand estradiol. Two cut-offs were used: LogRBA > -3 (“general binding”), and LogRBA > 0 (“strong binding”)<br />
	<br />		<br />

Line 216:		Line 216:
	=== Methods ===		=== Methods ===

	Probabilistic predictive models for all considered endpoints were developed using GALAS modeling methodology [4]. Each GALAS model consists of two parts:		Probabilistic predictive models for all considered endpoints were developed using GALAS modeling methodology [https://pubmed.ncbi.nlm.nih.gov/20814717]. Each GALAS model consists of two parts:
	# Global (baseline) model that reflects general trends in the property of interest. Baseline models were built using binomial PLS method based on fragmental descriptors.		# Global (baseline) model that reflects general trends in the property of interest. Baseline models were built using binomial PLS method based on fragmental descriptors.
	# Local corrections were applied to baseline predictions using a special similarity-based routine, after performing an analysis for the most similar compounds used in the training set. The local part of the model provides the basis for the calculation of the Reliability index (RI), a value ranging from 0 to 1 that provides a quantitative estimate of prediction accuracy.		# Local corrections were applied to baseline predictions using a special similarity-based routine, after performing an analysis for the most similar compounds used in the training set. The local part of the model provides the basis for the calculation of the Reliability index (RI), a value ranging from 0 to 1 that provides a quantitative estimate of prediction accuracy.
Line 224:		Line 224:
	=== Genotoxicity/Carcinogenicity Hazards ===		=== Genotoxicity/Carcinogenicity Hazards ===

	The knowledge-based expert system that identifies structural fragments potentially responsible for genotoxic effect is an extension of the previously described Ames mutagenicity hazards system [5]. The list of alerting groups was augmented with structural moieties that are frequently present in compounds tested positive in chromosomal damage assays, eucaryote gene mutation tests, as well as in carcinogens acting by non-genotoxic (epigenetic) mechanisms. The final list included 67 structural alerts, 14 of which represent epigenetic carcinogens (androgens, peroxisome proliferators, etc.).<br />		The knowledge-based expert system that identifies structural fragments potentially responsible for genotoxic effect is an extension of the previously described Ames mutagenicity hazards system [https://www.sciencedirect.com/science/article/abs/pii/S0378427408007054]. The list of alerting groups was augmented with structural moieties that are frequently present in compounds tested positive in chromosomal damage assays, eucaryote gene mutation tests, as well as in carcinogens acting by non-genotoxic (epigenetic) mechanisms. The final list included 67 structural alerts, 14 of which represent epigenetic carcinogens (androgens, peroxisome proliferators, etc.) [https://www.sciencedirect.com/science/article/abs/pii/S0378427411005492].<br />
	Overall, the expert system was able to detect 94% of mutagens in the Ames test DB and 90% of compounds labeled as potent carcinogens by FDA.<br />		Overall, the expert system was able to detect 94% of mutagens in the Ames test DB and 90% of compounds labeled as potent carcinogens by FDA.<br />

	The alert list is not limited to directly acting substructures, such as planar polycyclic arenes, aromatic amines, quinones, N-nitro and N-nitroso groups, but also includes various fragments that may undergo biotransformation to reactive intermediates. As an example, troglizatone, a thiazolidinedione class antidiabetic drug, was classified by the FDA as a potent carcinogen and has since been withdrawn from the USA market. The carcinogenic effect of this drug is mediated by several reactive metabolites. In human liver microsomes, the chromane ring of troglitazone is metabolized by CYP3A4 to form quinone and quinone-methide products. Furthermore, oxidative cleavage of thiazolidinedione ring results in a reactive sulfenic acid metabolite that also contains an isocyanate moiety [6].<br />		The alert list is not limited to directly acting substructures, such as planar polycyclic arenes, aromatic amines, quinones, N-nitro and N-nitroso groups, but also includes various fragments that may undergo biotransformation to reactive intermediates. As an example, troglizatone, a thiazolidinedione class antidiabetic drug, was classified by the FDA as a potent carcinogen and has since been withdrawn from the USA market. The carcinogenic effect of this drug is mediated by several reactive metabolites. In human liver microsomes, the chromane ring of troglitazone is metabolized by CYP3A4 to form quinone and quinone-methide products. Furthermore, oxidative cleavage of thiazolidinedione ring results in a reactive sulfenic acid metabolite that also contains an isocyanate moiety [https://pubmed.ncbi.nlm.nih.gov/20869346].<br />
	<br />		<br />

	=== References ===		<!--=== References ===
	[1] Matthews EJ et al. Regul Toxicol Pharmacol. 2007, 47, 115.<br>		[1] Matthews EJ et al. Regul Toxicol Pharmacol. 2007, 47, 115.<br>
	[2] Gold LS et al. Toxicol Sci. 2005, 85, 747.<br>		[2] Gold LS et al. Toxicol Sci. 2005, 85, 747.<br>
Line 237:		Line 237:
	[5] Didziapetris R et al. Toxicol Lett. 2008, 180, S152.<br>		[5] Didziapetris R et al. Toxicol Lett. 2008, 180, S152.<br>
	[6] Mansuy D & Dansette PM. Arch Biochem Biophys. 2011, 507, 1745.<br>		[6] Mansuy D & Dansette PM. Arch Biochem Biophys. 2011, 507, 1745.<br>
			-->
	</div>		</div>
	</div>		</div>

Kirilas: Fix copy/paste error from other module

2021-12-09T13:54:59Z

Fix copy/paste error from other module

← Older revision		Revision as of 13:54, 9 December 2021
Line 90:		Line 90:
	<br />		<br />

	~~===Calculated quantitative parameters===~~
	For modeling purposes experimental classification results of standard Draize test (not irritating, slightly irritating, irritating, etc.) have been transformed into a binary variable according to the following scheme: compounds producing at least moderate eye or skin irritation were considered positive and all the others - negative. The resulting probabilistic predictor estimates whether the analyzed compound is likely to act as as moderate or severe eye or skin irritant.
	{\| border="1" class="wikitable" style="margin: 1em auto 1em auto; text-align: left"		{\| border="1" class="wikitable" style="margin: 1em auto 1em auto; text-align: left"
	\|+ Table. A list of bioassays and dataset sizes included in ACD/Impurity Profiler.		\|+ Table. A list of bioassays and dataset sizes included in ACD/Impurity Profiler.

Kirilas: Updated for 2021.1 release

2021-07-28T08:58:58Z

Updated for 2021.1 release

@@ Line 11: / Line 11: @@
 * Endocrine disruption mechanisms (estrogen receptor binding)
 The impurities package offers probabilistic predictive models for 21 different endpoints that cover various mechanisms of hazardous activity presented above. These predictors are supplemented with a knowledge-based expert system that identifies potentially hazardous structural fragments that could be responsible for genotoxic and/or carcinogenic activity of the compound of interest.
 <br />
@@ Line 18: / Line 21: @@
 * Gain insight into the possible mechanisms of toxic effects
 * See a display of up to 5 similar structures with experimental results in relevant bioassays
 <br />
@@ Line 25: / Line 29: @@
 [[Image:Genotoxicity_impurity_profiling.png|center]]
 <br />
 # Hover over the name of an alert to highlight the alerting group on the structure of the molecule
 # The list of all alerting groups found in the molecule. Each alert is supplied with statistical data regarding distrubution of positive and negative compounds possessing this hazardous fragment in all considered databases, along with the respective z-scores. Z-scores show whether the presence of the fragment leads to a statistically significant increase in proportion of compounds with a positive test result for a particular assay. This information provides further evidence regarding the possible mechanisms of action.
@@ Line 47: / Line 52: @@
 The expert system contains a list of 67 alerting groups of toxicophores, 53 of which account for point mutational and/or clastogenic mechanisms of DNA damage, while the remaining 14 substructures detect carcinogens acting by non-genotoxic mechanisms. The expert system was able to recognize >94% of mutagens in ACD/Ames test database, and >90% of compounds marked as potent carcinogens in the FDA's OFAS Food-Additive Knowledgebase.
 <br />
 === Experimental Data ===

Kirilas at 10:01, 15 June 2017

2017-06-15T10:01:02Z

← Older revision		Revision as of 10:01, 15 June 2017
Line 35:		Line 35:


	<div class="mw-collapsible ~~mw-collapsed~~">		<div class="mw-collapsible">

	==Technical information==		==Technical information==

Kirilas: /* Overview */

2014-06-06T12:40:19Z

Overview

← Older revision		Revision as of 12:40, 6 June 2014
Line 4:		Line 4:
	Impurity profiling module is a result of the collaboration between ACD/Labs and FDA Center for Food Safety and Nutrition (CFSAN). Evaluation of genotoxic and/or carcinogenic potential is based on a battery of probabilistic models for bioassays reflecting different mechanisms of hazardous activity. A knowledge-based expert system identifies potentially hazardous structural fragments that could be responsible for carcinogenic activity of the test molecule.		Impurity profiling module is a result of the collaboration between ACD/Labs and FDA Center for Food Safety and Nutrition (CFSAN). Evaluation of genotoxic and/or carcinogenic potential is based on a battery of probabilistic models for bioassays reflecting different mechanisms of hazardous activity. A knowledge-based expert system identifies potentially hazardous structural fragments that could be responsible for carcinogenic activity of the test molecule.

	The toxicity predictions in the ~~Impurities Package~~ offer greater insight into the safety of impurities, providing detailed information on toxic endpoints, reflecting various mechanisms of hazardous activity including:		The toxicity predictions in the impurity profiling package offer greater insight into the safety of impurities, providing detailed information on toxic endpoints, reflecting various mechanisms of hazardous activity including:
	* Mutagenicity (Ames test, Mouse Lymphoma Assay, and other standard assays)		* Mutagenicity (Ames test, Mouse Lymphoma Assay, and other standard assays)
	* Clastogenicity (Micronucleus test, Chromosomal Aberrations)		* Clastogenicity (Micronucleus test, Chromosomal Aberrations)

Kirilas: /* Experimental Data */

2013-02-15T11:21:58Z

Experimental Data

Show changes

Kristina at 13:19, 29 May 2012

2012-05-29T13:19:02Z

← Older revision		Revision as of 13:19, 29 May 2012
Line 23:		Line 23:
	<br />		<br />

	[[Image:~~genotoxicity_impurity_profiling~~.png\|center]]		[[Image:Genotoxicity_impurity_profiling.png\|center]]
	<br />		<br />
	# Hover over the name of an alert to highlight the alerting group on the structure of the molecule		# Hover over the name of an alert to highlight the alerting group on the structure of the molecule
Line 58:		Line 58:
	'''Reproductive toxicity''': experimental data characterizing the potential for endocrine system disruption due to Estrogen receptor α binding were acquired from ChEMBL database [3]. Compounds were classified as binders/non-binders on the basis of their relative binding affinities (RBA) compared to reference ligand estradiol. Two cut-offs were used: LogRBA > -3 (“general binding”), and LogRBA > 0 (“strong binding”)<br />		'''Reproductive toxicity''': experimental data characterizing the potential for endocrine system disruption due to Estrogen receptor α binding were acquired from ChEMBL database [3]. Compounds were classified as binders/non-binders on the basis of their relative binding affinities (RBA) compared to reference ligand estradiol. Two cut-offs were used: LogRBA > -3 (“general binding”), and LogRBA > 0 (“strong binding”)<br />
	<br />		<br />
	[[File:~~genotoxicity_impurity_table~~.png\|center]]		[[File:Genotoxicity_impurity_table.png\|center]]
	<br />		<br />