Ames Test - Revision history

Kirilas at 10:45, 26 July 2023

2023-07-26T10:45:22Z

← Older revision		Revision as of 10:45, 26 July 2023
Line 16:		Line 16:
	* User-defined data can be added to the Self-training Library of the model for an instant improvement of accuracy and reliability of calculations for similar compounds.		* User-defined data can be added to the Self-training Library of the model for an instant improvement of accuracy and reliability of calculations for similar compounds.

			<!--
	<span style="color:red; font-weight: bold;">IMPORTANT NOTE:</span>		<span style="color:red; font-weight: bold;">IMPORTANT NOTE:</span>

Line 21:		Line 22:

	In case of a new installation, or an upgrade over ACD/Percepta v. 2018, the new library should be selected automatically with no further action required.		In case of a new installation, or an upgrade over ACD/Percepta v. 2018, the new library should be selected automatically with no further action required.
			-->
	== Interface ==		== Interface ==
	<br />		<br />

Kirilas: Fixed broken links

2022-10-20T08:24:18Z

Fixed broken links

← Older revision		Revision as of 08:24, 20 October 2022
Line 69:		Line 69:
	Classification of compounds as genotoxic or non-genotoxic performed by ACD/Percepta is highly accurate: model validation results demonstrate that less than 5% of test set compounds (~20% of the overall data set) are mispredicted if inconclusive (0.2 < p < 0.8) and unreliable (''RI'' < 0.3) predictions are not considered. If probability threshold value of 0.5 is used the number of mispredicts still only slightly exceeds 10% of the test set. The percentages of misclassified compounds and inconclusive predictions (calculated probabilities falling in the range 0.2-0.8) also decrease significantly if only predictions of moderate and high reliability (''RI'' > 0.5) are considered (see Table below).		Classification of compounds as genotoxic or non-genotoxic performed by ACD/Percepta is highly accurate: model validation results demonstrate that less than 5% of test set compounds (~20% of the overall data set) are mispredicted if inconclusive (0.2 < p < 0.8) and unreliable (''RI'' < 0.3) predictions are not considered. If probability threshold value of 0.5 is used the number of mispredicts still only slightly exceeds 10% of the test set. The percentages of misclassified compounds and inconclusive predictions (calculated probabilities falling in the range 0.2-0.8) also decrease significantly if only predictions of moderate and high reliability (''RI'' > 0.5) are considered (see Table below).

	For a competitive evaluation of ACD/Genotoxicity predictor please refer to an Application Note [http://~~www~~.acdlabs.com/~~download/app/admet~~/1005_Tox_Genotoxicity.pdf].		For a competitive evaluation of ACD/Genotoxicity predictor please refer to an Application Note [http://perceptahelp.acdlabs.com/docs/1005_Tox_Genotoxicity.pdf].

	Futhermore, more than 90% of genotoxic compounds from the entire data set are covered by the list of hazardous substructures used in Genotoxicity Hazards module.		Futhermore, more than 90% of genotoxic compounds from the entire data set are covered by the list of hazardous substructures used in Genotoxicity Hazards module.

Kirilas: Edited new library note

2019-09-24T12:20:20Z

Edited new library note

← Older revision		Revision as of 12:20, 24 September 2019
Line 18:		Line 18:
	<span style="color:red; font-weight: bold;">IMPORTANT NOTE:</span>		<span style="color:red; font-weight: bold;">IMPORTANT NOTE:</span>

	If you installed ACD/Percepta ~~v. 2018~~ as an ~~in-place~~ upgrade over ~~the~~ previous version, ~~Ames Test module~~ will ~~continue using the same set~~ of Self-training libraries ~~that was configured~~ in ~~your previous installation~~. ~~To take advantage of~~ the new, significantly extended built-in library that ~~comes with the current version~~ of the ~~software~~, you ~~will~~ need to click "Configure" and manually select the ~~folowing~~ entry: ''AMES Test v. 1.3 (read-only)''.		If you installed ACD/Percepta as an upgrade over a previous version, the program will attempt to preserve any custom configuration of Self-training libraries used in Ames Test module. If you are upgrading from a version earlier than v. 2018, this configuration will not include the new, significantly extended built-in library that was introduced in 2018 release. In this case, to take advantage of the new library, you may need to click "Configure" and manually select the following entry: ''AMES Test v. 1.3 (read-only)''.

	In case of a ~~clean~~ installation, ~~this~~ library is selected automatically~~, and~~ no further action is required.		In case of a new installation, or an upgrade over ACD/Percepta v. 2018, the new library should be selected automatically with no further action required.
	== Interface ==		== Interface ==
	<br />		<br />

Kirilas at 10:54, 30 October 2018

2018-10-30T10:54:16Z

← Older revision		Revision as of 10:54, 30 October 2018
Line 15:		Line 15:
	* Overall experimental Ames test results with references for up to 5 similar structures from the training set are displayed along with each prediction. Additionally, access to a fully browsable Ames Test DB is available, providing information about individual studies conducted with each compound corresponding to various bacterial strains tested, presence or absence of metabolic activation, as well as other experimental conditions.		* Overall experimental Ames test results with references for up to 5 similar structures from the training set are displayed along with each prediction. Additionally, access to a fully browsable Ames Test DB is available, providing information about individual studies conducted with each compound corresponding to various bacterial strains tested, presence or absence of metabolic activation, as well as other experimental conditions.
	* User-defined data can be added to the Self-training Library of the model for an instant improvement of accuracy and reliability of calculations for similar compounds.		* User-defined data can be added to the Self-training Library of the model for an instant improvement of accuracy and reliability of calculations for similar compounds.
	~~<br />~~

			<span style="color:red; font-weight: bold;">IMPORTANT NOTE:</span>

			If you installed ACD/Percepta v. 2018 as an in-place upgrade over the previous version, Ames Test module will continue using the same set of Self-training libraries that was configured in your previous installation. To take advantage of the new, significantly extended built-in library that comes with the current version of the software, you will need to click "Configure" and manually select the folowing entry: ''AMES Test v. 1.3 (read-only)''.

			In case of a clean installation, this library is selected automatically, and no further action is required.
	== Interface ==		== Interface ==
	<br />		<br />

Kirilas: /* Technical information */

2018-08-08T09:54:08Z

Technical information

← Older revision		Revision as of 09:54, 8 August 2018
Line 90:		Line 90:
	! Genotoxic \|\| style="background:#FFE1E1" \| 23 (2.1%) \|\| style="background:#E1FFE1" \| 786 (70.4%) \|\| style="background:#FFE1E1" \| 11 (0.9%) \|\| style="background:#FFFFAF" \| 38 (3.4%) \|\| style="background:#E1FFE1" \| 760 (68.0%)		! Genotoxic \|\| style="background:#FFE1E1" \| 23 (2.1%) \|\| style="background:#E1FFE1" \| 786 (70.4%) \|\| style="background:#FFE1E1" \| 11 (0.9%) \|\| style="background:#FFFFAF" \| 38 (3.4%) \|\| style="background:#E1FFE1" \| 760 (68.0%)
	\|}		\|}

			The statistics above apply to ACD/Mutagenicity prediction algorithm with built-in "Ames Test v. 1.2" self-training library. The current version uses an updated "Ames Test v. 1.3" library that has been expanded with experimental data from approximately 1700 compounds obtained from regulatory reports (a 20% increase in size). Inclusion of the new data has resulted in improved prediction accuracy and reliability for novel marketed drugs, and for impurities and degradants of drug-like compounds:

			[[File:Ames_Test_v2017_to_v2018.png\|frame\|none\|Improvements in sensitivity, specificity, and accuracy of AMES test predictions for the new compounds used in expansion of the internal database in v2018.1]]

Kirilas at 10:00, 15 June 2017

2017-06-15T10:00:08Z

← Older revision		Revision as of 10:00, 15 June 2017
Line 31:		Line 31:


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

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

Kirilas: /* Model features & prediction accuracy */

2013-02-15T10:03:24Z

Model features & prediction accuracy

← Older revision		Revision as of 10:03, 15 February 2013
Line 64:		Line 64:

	Classification of compounds as genotoxic or non-genotoxic performed by ACD/Percepta is highly accurate: model validation results demonstrate that less than 5% of test set compounds (~20% of the overall data set) are mispredicted if inconclusive (0.2 < p < 0.8) and unreliable (''RI'' < 0.3) predictions are not considered. If probability threshold value of 0.5 is used the number of mispredicts still only slightly exceeds 10% of the test set. The percentages of misclassified compounds and inconclusive predictions (calculated probabilities falling in the range 0.2-0.8) also decrease significantly if only predictions of moderate and high reliability (''RI'' > 0.5) are considered (see Table below).		Classification of compounds as genotoxic or non-genotoxic performed by ACD/Percepta is highly accurate: model validation results demonstrate that less than 5% of test set compounds (~20% of the overall data set) are mispredicted if inconclusive (0.2 < p < 0.8) and unreliable (''RI'' < 0.3) predictions are not considered. If probability threshold value of 0.5 is used the number of mispredicts still only slightly exceeds 10% of the test set. The percentages of misclassified compounds and inconclusive predictions (calculated probabilities falling in the range 0.2-0.8) also decrease significantly if only predictions of moderate and high reliability (''RI'' > 0.5) are considered (see Table below).

	For a competitive evaluation of ACD/Genotoxicity predictor please refer to an Application Note [http://www.acdlabs.com/download/app/admet/1005_Tox_Genotoxicity.pdf].		For a competitive evaluation of ACD/Genotoxicity predictor please refer to an Application Note [http://www.acdlabs.com/download/app/admet/1005_Tox_Genotoxicity.pdf].

	Futhermore, more than 90% of genotoxic compounds from the entire data set are covered by the list of hazardous substructures used in Genotoxicity Hazards module.		Futhermore, more than 90% of genotoxic compounds from the entire data set are covered by the list of hazardous substructures used in Genotoxicity Hazards module.

Kirilas: /* Model features & prediction accuracy */

2013-02-15T10:02:34Z

Model features & prediction accuracy

← Older revision		Revision as of 10:02, 15 February 2013
Line 64:		Line 64:

	Classification of compounds as genotoxic or non-genotoxic performed by ACD/Percepta is highly accurate: model validation results demonstrate that less than 5% of test set compounds (~20% of the overall data set) are mispredicted if inconclusive (0.2 < p < 0.8) and unreliable (''RI'' < 0.3) predictions are not considered. If probability threshold value of 0.5 is used the number of mispredicts still only slightly exceeds 10% of the test set. The percentages of misclassified compounds and inconclusive predictions (calculated probabilities falling in the range 0.2-0.8) also decrease significantly if only predictions of moderate and high reliability (''RI'' > 0.5) are considered (see Table below).		Classification of compounds as genotoxic or non-genotoxic performed by ACD/Percepta is highly accurate: model validation results demonstrate that less than 5% of test set compounds (~20% of the overall data set) are mispredicted if inconclusive (0.2 < p < 0.8) and unreliable (''RI'' < 0.3) predictions are not considered. If probability threshold value of 0.5 is used the number of mispredicts still only slightly exceeds 10% of the test set. The percentages of misclassified compounds and inconclusive predictions (calculated probabilities falling in the range 0.2-0.8) also decrease significantly if only predictions of moderate and high reliability (''RI'' > 0.5) are considered (see Table below).
			For a competitive evaluation of ACD/Genotoxicity predictor please refer to an Application Note [http://www.acdlabs.com/download/app/admet/1005_Tox_Genotoxicity.pdf].
	Futhermore, more than 90% of genotoxic compounds from the entire data set are covered by the list of hazardous substructures used in Genotoxicity Hazards module.		Futhermore, more than 90% of genotoxic compounds from the entire data set are covered by the list of hazardous substructures used in Genotoxicity Hazards module.

Kirilas: /* Model features & prediction accuracy */

2013-02-14T10:55:31Z

Model features & prediction accuracy

@@ Line 46: / Line 46: @@
 ===Data interpretation and assignment of qualitative categories===
 In the Ames Genotoxicity database, compounds were classified as Ames positive if it demonstrated clear positive results in at least one tested strain with or without metabolic activation. Compounds that did not increase the frequency of revertants in all tested strains were considered safe (Ames negative). Some chemicals that consistently exhibited weak mutagenic activity were marked weakly positive while in those cases when the results of different studies were discrepant the corresponding compounds were labeled inconclusive.
 ===Model features & prediction accuracy===
-Classification of compounds as genotoxic or non-genotoxic performed by Tox Boxes is highly accurate: model validation results demonstrate that less than 5% of test set compounds (~20% of the overall data set) are mispredicted if inconclusive (0.2 < p < 0.8) and unreliable (RI < 0.3) predictions are not considered. If probability threshold value of 0.5 is used the number of mispredicts still only slightly exceeds 10% of the test set. The percentages of misclassified compounds and inconclusive predictions (calculated probabilities falling in the range 0.2-0.8) also decrease significantly if only predictions of moderate and high reliability (RI > 0.5) are considered (see Table below).
+The predictive model of Ames mutagenicity was derived using GALAS (Global, Adjusted Locally According to Similarity) modeling methodology (please refer to [http://www.ncbi.nlm.nih.gov/pubmed/20373217] for more details).
 Futhermore, more than 90% of genotoxic compounds from the entire data set are covered by the list of hazardous substructures used in Genotoxicity Hazards module.
@@ Line 81: / Line 88: @@
 ! Genotoxic || style="background:#FFE1E1" | 23 (2.1%) || style="background:#E1FFE1" | 786 (70.4%) || style="background:#FFE1E1" | 11 (0.9%) || style="background:#FFFFAF" | 38 (3.4%) || style="background:#E1FFE1" | 760 (68.0%)
 |}

Kristina: /* Interface */

2012-05-29T13:18:01Z

Interface

← Older revision		Revision as of 13:18, 29 May 2012
Line 20:		Line 20:
	<br />		<br />

	[[Image:~~genotoxicity_ames_test~~.png\|center]]		[[Image:Genotoxicity_ames_test.png\|center]]
	<br />		<br />