Lhasa Limited shared knowledge shared progress

Negative Predictions for Bacterial Mutagenicity

Derek Nexus contains expert-derived functionality to provide negative predictions for bacterial in vitro mutagenicity. It has been designed to support your post-processing workflows, for example by highlighting structural features for further expert assessment as part of ICH M7 compliant submissions.

For further information regarding Negative Predictions and how this has been implemented within Derek Nexus, please take a look at a paper written by Richard Williams et al. in Regulatory Toxicology and Pharmacology: It's difficult, but important to make negative predictions. In addition a recorded webinar by Richard Williams highlighting the science and functionality of Negative Predictions with Derek Nexus can also be viewed.  

Functionality

The new functionality further evaluates those compounds which do not fire any bacterial in vitro (Ames test) mutagenicity alerts in Derek Nexus. The query compound is compared to a Lhasa reference set of Ames test data, producing the following outcomes:

  • In compounds where all features in the molecule are found in accurately classified compounds from the reference set, a negative prediction is displayed.
  • For those query compounds where features in the molecule are found in non-alerting mutagens in the Lhasa reference set, the prediction remains negative and the misclassified1 features are highlighted to enable the negative prediction to be verified by expert assessment.
  • In cases where features in the molecule are not found in the Lhasa reference set, the prediction remains negative and the unclassified2 features are highlighted to enable the negative prediction to be verified by expert assessment. 

Lhasa Ames test reference set

The Lhasa reference set of Ames test data referred to above, is composed of six Ames test datasets.

Data SetNumber of CompoundsNumber of PositivesNumber of NegativesNumber of Equivocal or Inconclusive
Vitic NTP 2001 591 1235 175
Hansen 6512 3505 3008 0
FDA CFSAN 8421 4300 4115 6
ISSSTY 7363 3574 2665 1124
CGX* 718 360 343 15
Marketed Pharmaceuticals 554 40 494 20

 

These datasets were combined and curated (including the removal of duplicates with non-equivalent results and compounds with equivocal or inconclusive results) to produce the Lhasa Ames test references set with the following composition:

Data SetNumber of CompoundsNumber of PositivesNumber of NegativesNumber of Equivocal or Inconclusive
Lhasa Ames test reference set 10243 5177 5066 0

 

  1. Misclassified features are those that have been derived from non-alerting mutagens in the Lhasa Ames test reference set
  2. Unclassified features are those that have not been found in the Lhasa Ames test reference set

* CGX database can be downloaded from the Lhasa Limited CGX webpage

 

Benefits

  • Combines Derek mutagenicity alerts with functionality to generate negative predictions
  • Fits seamlessly into the Derek Nexus user interface and reports
  • Supports expert review and assessment [Dobo et all 2012, Sutter et al 2013].
  • No "Nothing to report" or "Out of Domain"
  • For bacterial mutagenicity predictions only

Workings

The work flow used by Derek Nexus to provide a prediction on bacterial mutagenicity is described below.

Workflow used by Derek Nexus for negative predictions

Performance metrics

3 proprietary datasets were used to assess the accuracy of the negative predictions, the results and composition of each can be seen below. 

Describing the composition of positive and negative ames data for the entire dataset

 

Vitic Intermediates data set No. of cpds. Proportion in data set Negative predictivity
All compounds in analysis 854 100% n/c
Alerting compounds 361 42.3% n/c
Non-alerting compounds Inactive 464 54.3% 86.0%
Inactive with misclassified features 20 2.3% 80.0%
Inactive with unclassified features 9 1.1% 66.7%
Inactive with unclassified and misclassified features 0 0% n/c

 

Proprietary data set 1 No. of cpds. Proportion in data set Negative predictivity
All compounds in analysis 364 100% n/c
Alerting compounds 39 10.7% n/c
Non-alerting compounds Inactive 280 76.9% 94.3%
Inactive with misclassified features 29 8.0% 86.2%
Inactive with unclassified features 15 4.1% 100%
Inactive with unclassified and misclassified features 1 n/c n/c

 

 Proprietary data set 2 No. of cpds. Proportion in data set Negative predictivity
All compounds in analysis 503 100% n/c
Alerting compounds 87 17.3% n/c
Non-alerting compounds Inactive 372 74.0% 87.4%
Inactive with misclassified features 13 2.6% 84.6%
Inactive with unclassified features 31 6.2% 93.5%
Inactive with unclassified and misclassified features 0 0% n/c

 

Contact Us

© 2017 Lhasa Limited | Registered office: Granary Wharf House, 2 Canal Wharf, Leeds, LS11 5PS, UK Tel: +44 (0)113 394 6020
VAT number 396 8737 77 | Lhasa Limited is registered as a charity (290866)| Company Registration Number 01765239 (England and Wales).

QuestionPro supports sample survey questions such as multiple choice, drop-down menu, likert-scale, semantic differential, matrix, constant sum, drag-and-drop, slider-scale, net-promoter scale, and many more question types.