AKSW - Projects/DL-Learner/Carcinogenesis

2008-03-25 16:34:50

Comparison for /Projects / DL- Learner / Carcinogenesis of 2008-03-25 16:34:50 and 2008-04-15 12:20:02

Additions:

or the ames test of the compound is positive and there are at least five atom bonds which are not of bond type 7.

Deletions:

or the ames test of the compound is positive and the there are at least five atom bonds which are not of bond type 7.

2008-03-24 10:51:51

Comparison for /Projects / DL- Learner / Carcinogenesis of 2008-03-24 10:51:51 and 2008-03-25 16:34:50

Additions:

Deletions:

2008-03-20 19:21:38

Comparison for /Projects / DL- Learner / Carcinogenesis of 2008-03-20 19:21:38 and 2008-03-24 10:51:51

Additions:

We used DL-Learner to learn general models for substances causing cancer. The most sensible parameter of DL-Learner in this case is noise (bounding the minimum acceptable training set accuracy of the learned definition). We varied the noise parameter starting from 35% downwards in one percent steps and finally determined 32% as the setting, which minimizes the 10 fold cross validation error. The following table summarizes the accuracy of the learned definitions in comparison with other approaches using the same background knowledge (many of those use Aleph – a state of the art Inductive Logic Programming system – as their basis):
As evident from the table, DL-Learner has the highest cross validation accuracy of the listed approaches. For all approaches, where the standard deviation was given in the articles, we calculated whether the difference in accuracy is statistically significant using a t-test with a confidence interval of 95%. Indeed, in all those cases the difference turned out to be statistically significant. Note, that the second and third table entry combine different learned rules to obtain a classifier, which increases accuracy but reduces readability of the result. The average runtime of DL-Learner for each of the ten folds was 178 seconds on a 2.2Ghz Dual core machine with 2GB RAM. Additional 36 seconds were needed to read in the ontology (using the OWL API) and prepare it for the approximate reasoner built into DL-Learner. Overall, DL-Learner reaches a similar accuracy as those approaches involving (possible task-specific) additional preprocessing steps.

Deletions:

We used DL-Learner to learn general models for substances causing cancer. The most sensible parameter of DL-Learner in this case is noise (bounding the minimum acceptable training set accuracy of the learned definition). We varied the noise parameter starting from 35% downwards in one percent steps and finally determined 32% as the setting, which minimizes the 10 fold cross validation error. The following table summarizes the accuracy of the learned definitions in comparison with other approaches (many of those use Aleph – a state of the art Inductive Logic Programming system – as their basis):
As evident from the table, DL-Learner has the highest cross validation accuracy of the listed approaches. For all approaches, where the standard deviation was given in the articles, we calculated whether the difference in accuracy is statistically significant using a t-test with a confidence interval of 95%. Indeed, in all those cases the difference turned out to be statistically significant. Note, that the second and third table entry combine different learned rules to obtain a classifier, which increases accuracy but reduces readability of the result. The average runtime of DL-Learner for each of the ten folds was 178 seconds on a 2.2Ghz Dual core machine with 2GB RAM. Additional 36 seconds were needed to read in the ontology (using the OWL API) and prepare it for the approximate reasoner built into DL-Learner.

2008-03-20 19:20:39

Comparison for /Projects / DL- Learner / Carcinogenesis of 2008-03-20 19:20:39 and 2008-03-20 19:21:38

Additions:

For each compound, the chemical structure, structural indicators, and the results of short-term assays were made available. They can be downloaded at the website of the Oxford University Machine Learning Group. The data was made available in Prolog format. It is well known that not every logic program can be converted to an OWL knowledge base and vice versa. However, for the carcinogenesis problem, such a mapping is possible. DL-Learner contains a Prolog parser, which was used to read in the data. Using a set of mapping rules, the Prolog files were converted into an OWL ontology, which can be downloaded here and viewed using e.g. Proteg� or OntoWiki. During the transformation process almost no knowledge was lost or added. The resulting ontology contains 142 classes, 19 properties, 22373 instances, and 74567 triples.

Deletions:

For each compound, the chemical structure, structural indicators, and the results of short-term assays were made available. They can be downloaded at the website of the Oxford University Machine Learning Group. The data was made available in Prolog format. It is well known that not every logic program can be converted to an OWL knowledge base and vice versa. However, for the carcinogenesis problem, such a mapping is possible. DL-Learner contains a Prolog parser, which was used to read in the data. Using a set of mapping rules, the Prolog files were converted into an OWL ontology, which can be downloaded here and can be viewed using e.g. Proteg� or OntoWiki. During the transformation process almost no knowledge was lost or added. The resulting ontology contains 142 classes, 19 properties, 22373 instances, and 74567 triples.

2008-03-20 19:09:40

Comparison for /Projects / DL- Learner / Carcinogenesis of 2008-03-20 19:09:40 and 2008-03-20 19:20:39

Additions:

There have been several approaches by both – human and machines – to predict carcinogenicity. It has been shown in [6] that the performance of machine derived models for carcinogenicity can be equal to human experts. Classifying chemicals is a massive challenge, due to the high number and diversity of elements, structures, and tests involved in the problem. As outlined in several journal articles in the area of bioinformatics [1,3,7], reliable carcinogenesis prediction has not been achieved yet, but Artificial Intelligence approaches were able to learn new or confirm existing toxicological knowledge. The use of carcinogenicity models, which can be derived by humans and AI approaches, for setting priorities in the experiments of the The U.S. National Toxicology Program (NTP) has been succesful [1].

Deletions:

There have been several approaches by both – human and machines – to predict carcinogenicity. It has been shown in [6] that the performance of machine derived models for carcinogenicity can be equal to human experts in carcinogenicity prediction. Classifying chemicals is a massive challenge, due to the high number and diversity of elements, structures, and tests involved in the problem. As outlined in several journal articles in the area of bioinformatics [1,3,7], reliable carcinogenesis prediction has not been achieved yet, but Artificial Intelligence approaches were able to learn new or confirm existing toxicological knowledge. The use of carcinogenicity models, which can be derived by humans and AI approaches, for setting priorities in the experiments of the The U.S. National Toxicology Program (NTP) has been succesful [1].

2008-03-20 12:57:35

Comparison for /Projects / DL- Learner / Carcinogenesis of 2008-03-20 12:57:35 and 2008-03-20 19:09:40

Additions:

As evident from the table, DL-Learner has the highest cross validation accuracy of the listed approaches. For all approaches, where the standard deviation was given in the articles, we calculated whether the difference in accuracy is statistically significant using a t-test with a confidence interval of 95%. Indeed, in all those cases the difference turned out to be statistically significant. Note, that the second and third table entry combine different learned rules to obtain a classifier, which increases accuracy but reduces readability of the result. The average runtime of DL-Learner for each of the ten folds was 178 seconds on a 2.2Ghz Dual core machine with 2GB RAM. Additional 36 seconds were needed to read in the ontology (using the OWL API) and prepare it for the approximate reasoner built into DL-Learner.

Deletions:

As evident from the table, DL-Learner has the highest cross validation accuracy of the listed approaches. For all approaches, where the standard deviation was given in the articles, we calculated whether the difference in accuracy is statistically significant using a t-test with a confidence interval of 95%. Indeed, in all those cases the difference turned out to be statistically significant. Note, that the second and third table entry combine different learned rules to obtain a classifier, which increases accuracy but reduces readability of the result. The average runtime of DL-Learner for the ten folds was 178 seconds on a 2.2Ghz Dual core machine with 2GB RAM. Additional 36 seconds were needed to read in the ontology (using the OWL API) and prepare it for the approximate reasoner built into DL-Learner.

2008-03-20 12:19:42

Comparison for /Projects / DL- Learner / Carcinogenesis of 2008-03-20 12:19:42 and 2008-03-20 12:57:35

Additions:

... and it has at least three Halide – excluding Halide10 – structures
or the ames test of the compound is positive and the there are at least five atom bonds which are not of bond type 7.

Deletions:

... it has at least three Halide – excluding Halide10 – structures
or the ames test of the compound is positive and the there are at least five atom bonds which are not of bond type 7.

2008-03-20 12:18:07

Comparison for /Projects / DL- Learner / Carcinogenesis of 2008-03-20 12:18:07 and 2008-03-20 12:19:42

Additions:

Deletions:

2008-03-20 12:13:40

Comparison for /Projects / DL- Learner / Carcinogenesis of 2008-03-20 12:13:40 and 2008-03-20 12:18:07

Additions:

Deletions: