references added

4 files changed, 394 insertions, 205 deletions

diff --git a/loael.Rmd b/loael.Rmd
index 8916685..2a32482 100644
--- a/loael.Rmd
+++ b/loael.Rmd
@@ -37,32 +37,31 @@ Introduction
 
 Relying on standard animal toxicological testing for chemical hazard
 identification and characterization is increasingly questioned on both
-scientific and ethical grounds. In addition, it appears obvious that
-from a resource perspective, the capacity of standard toxicology to
-address the safety of thousands of untested chemicals (Fowler et al.,
-2011) to which human may be exposed is very limited. It has also been
-recognized that getting rapid insight on toxicity of chemicals in case
-of emergency safety incidents or for early prioritization in research
-and development (safety by design) is a big challenge mainly because of
-the time and cost constraints associated with the generation of relevant
-animal data. In this context, alternative approaches to obtain timely
-and fit-for-purpose toxicological information are being developed.
-Amongst others, non-testing, structure-activity based *in silico*
-toxicology methods (also called computational toxicology) are considered
-highly promising. Importantly, they are raising more and more interests
-and getting increased acceptance in various regulatory (e.g. ECHA, 2008;
-EFSA, 2016, 2014; Health Canada, 2016; OECD, 2015) and industrial (e.g.
-Stanton and Kruszewski, 2016; Lo Piparo et al., 2011) frameworks.
+scientific and ethical grounds. In addition, it appears obvious that from
+a resource perspective, the capacity of standard toxicology to address the
+safety of thousands of untested chemicals [@Fowler2011] to which human may be
+exposed is very limited. It has also been recognized that getting rapid insight
+on toxicity of chemicals in case of emergency safety incidents or for early
+prioritization in research and development (safety by design) is a big
+challenge mainly because of the time and cost constraints associated with the
+generation of relevant animal data. In this context, alternative approaches to
+obtain timely and fit-for-purpose toxicological information are being
+developed. Amongst others, non-testing, structure-activity based *in silico*
+toxicology methods (also called computational toxicology) are considered highly
+promising. Importantly, they are raising more and more interests
+and getting increased acceptance in various regulatory (e.g.
+[@ECHA2008, @EFSA2016, @EFSA2014, @HealthCanada2016, @OECD2015]) and industrial (e.g.
+[@Stanton2016, @LoPiparo2011]) frameworks.
 
 For a long time already, computational methods have been an integral
 part of pharmaceutical discovery pipelines, while in chemical food
-safety their actual potentials emerged only recently (Lo Piparo et al.,
-2011). In this later field, an application considered critical is in the
+safety their actual potentials emerged only recently [@LoPiparo2011].
+In this later field, an application considered critical is in the
 establishment of levels of safety concern in order to rapidly and
 efficiently manage toxicologically uncharacterized chemicals identified
 in food. This requires a risk-based approach to benchmark exposure with
-a quantitative value of toxicity relevant for risk assessment (Schilter
-et al., 2014a). Since most of the time chemical food safety deals with
+a quantitative value of toxicity relevant for risk assessment [@Schilter2014].
+Since most of the time chemical food safety deals with
 life-long exposures to relatively low levels of chemicals, and because
 long-term toxicity studies are often the most sensitive in food
 toxicology databases, predicting chronic toxicity is of prime
@@ -136,13 +135,11 @@ The Nestlé database can be obtained from the following GitHub links: [original
 ### Swiss Food Safety and Veterinary Office (FSVO) database
 
 Publicly available data from pesticide evaluations of chronic rat
-toxicity studies from the European Food Safety Authority (EFSA) (EFSA,
-2014), the Joint FAO/WHO Meeting on Pesticide Residues (JMPR) (WHO,
-2011) and the US EPA (US EPA, 2011) were compiled to form the
+toxicity studies from the European Food Safety Authority (EFSA) [@EFSA2014], the Joint FAO/WHO Meeting on Pesticide Residues (JMPR) [@WHO2011] and the US EPA [@EPA2011] were compiled to form the
 FSVO-database. Only studies providing both an experimental NOAEL and an
 experimental LOAEL were included. The LOAELs were taken as they were
 reported in the evaluations. Further details on the database are
-described elsewhere (Zarn et al., 2011; Zarn et al., 2013). The
+described elsewhere [@Zarn2011, @Zarn2013]. The
 FSVO-database consists of `r length(s$SMILES)` rat LOAEL values for `r length(unique(s$SMILES))` unique chemical
 structures. It can be obtained from the following GitHub links:
 
@@ -172,8 +169,8 @@ dataset](https://github.com/opentox/loael-paper/blob/submission/data/test_log10.
 contains data from compounds that occur in both databases. LOAEL values equal
 at five significant digits were considered as duplicates originating from the
 same study/publication and only one instance was kept in the test dataset. The
-test dataset has `r length(t$SMILES)` LOAEL values for `r
-length(unique(t$SMILES))` unique chemical structures and was used for
+test dataset has `r length(t$SMILES)` LOAEL values for `r length(unique(t$SMILES))`
+unique chemical structures and was used for
 
 - evaluating experimental variability
 - comparing model predictions with experimental variability.
@@ -439,8 +436,8 @@ c.mg = read.csv("data/all_mg_dup.csv",header=T)
 c.mg$sd <- ave(c.mg$LOAEL,c.mg$SMILES,FUN=sd)
 ```
 
-The Nestlé database has `r length(m$SMILES)` LOAEL values for `r
-length(levels(m$SMILES))` unique structures, `r m.dupnr` compounds have
+The Nestlé database has `r length(m$SMILES)` LOAEL values for
+`r length(levels(m$SMILES))` unique structures, `r m.dupnr` compounds have
 multiple measurements with a mean standard deviation (-log10 transformed
 values) of `r round(mean(m.dup$sd),2)` (`r round(mean(10^(-1*m.mg$sd)),2)`
 mg/kg_bw/day, `r round(mean(10^(-1*m.dup$sd)),2)` mmol/kg_bw/day)
@@ -483,8 +480,8 @@ datasets. As both datasets contain duplicates medians were used for the
 correlation plot and statistics. It should be kept in mind that the aggregation of duplicated
 measurements into a single median value hides a substantial portion of the
 experimental variability.  Correlation analysis shows a significant (p-value < 2.2e-16)
-correlation between the experimental data in both datasets with r\^2: `r
-round(median.r.square,2)`, RMSE: `r round(median.rmse,2)`
+correlation between the experimental data in both datasets with r\^2:
+`r round(median.r.square,2)`, RMSE: `r round(median.rmse,2)`
 
 ![Correlation of median LOAEL values from Nestlé and FSVO databases. Data with
   identical values in both databases was removed from
@@ -510,8 +507,8 @@ correct_predictions = length(training$SMILES)-incorrect_predictions
 In order to compare the performance of *in silico* read across models with
 experimental variability we are using compounds that occur in both datasets as
 a test set (`r  length(t$SMILES)` measurements, `r  length(unique(t$SMILES))`
-compounds). `lazar` read across predictions were obtained for `r
-length(unique(t$SMILES))` compounds, `r  length(unique(t$SMILES)) - length(training$SMILES)`
+compounds). `lazar` read across predictions were obtained for
+`r length(unique(t$SMILES))` compounds, `r  length(unique(t$SMILES)) - length(training$SMILES)`
 predictions failed, because no similar compounds were found in the training
 data (i.e. they were not covered by the applicability domain of the training
 data).
@@ -632,11 +629,11 @@ It is currently acknowledged that there is a strong need for
 toxicological information on the multiple thousands of chemicals to
 which human may be exposed through food. These include for examples many
 chemicals in commerce, which could potentially find their way into food
-(Stanton and Kruszewski, 2016; Fowler et al., 2011), but also substances
-migrating from food contact materials (Grob et al., 2006), chemicals
-generated over food processing (Cottererill et al., 2008), environmental
-contaminants as well as inherent plant toxicants (Schilter et al.,
-2014b). For the vast majority of these chemicals, no toxicological data
+[@Stanton2016, @Fowler2011], but also substances
+migrating from food contact materials [@Grob2006], chemicals
+generated over food processing [@Cotterill2008], environmental
+contaminants as well as inherent plant toxicants [@Schilter2013].
+For the vast majority of these chemicals, no toxicological data
 is available and consequently insight on their potential health risks is
 very difficult to obtain. It is recognized that testing all of them in
 standard animal studies is neither feasible from a resource perspective
@@ -650,7 +647,7 @@ toxicology is thought to play an important role for that.
 In order to establish the level of safety concern of food chemicals
 toxicologically not characterized, a methodology mimicking the process
 of chemical risk assessment, and supported by computational toxicology,
-was proposed (Schilter et al., 2014a). It is based on the calculation of
+was proposed [@Schilter2014]. It is based on the calculation of
 margins of exposure (MoE) between predicted values of toxicity and
 exposure estimates. The level of safety concern of a chemical is then
 determined by the size of the MoE and its suitability to cover the
@@ -665,7 +662,7 @@ carcinogenic potency were developed. In these models, substances in the
 training dataset similar to the query compounds are automatically
 identified and used to derive a quantitative TD50 value. The errors
 observed in these models were within the published estimation of
-experimental variability (Lo Piparo, et al., 2014). In the present
+experimental variability [@LoPiparo2014]. In the present
 study, a similar approach was applied to build models generating
 quantitative predictions of long-term toxicity. Two databases compiling
 chronic oral rat lowest adverse effect levels (LOAEL) as endpoint were
@@ -726,7 +723,7 @@ than chonic studies should be studied. It is likely that more substances
 reflecting a wider chemical domain may be available. To predict such
 shorter duration endpoints would also be valuable for chronic toxicy
 since evidence suggest that exposure duration has little impact on the
-levels of NOAELs/LOAELs (Zarn et al., 2011, 2013).
+levels of NOAELs/LOAELs [@Zarn2011, @Zarn2013].
 
 <!--
 Elena + Benoit
diff --git a/loael.md b/loael.md
index 60ec21d..f2a967c 100644
--- a/loael.md
+++ b/loael.md
@@ -29,32 +29,31 @@ Introduction
 
 Relying on standard animal toxicological testing for chemical hazard
 identification and characterization is increasingly questioned on both
-scientific and ethical grounds. In addition, it appears obvious that
-from a resource perspective, the capacity of standard toxicology to
-address the safety of thousands of untested chemicals (Fowler et al.,
-2011) to which human may be exposed is very limited. It has also been
-recognized that getting rapid insight on toxicity of chemicals in case
-of emergency safety incidents or for early prioritization in research
-and development (safety by design) is a big challenge mainly because of
-the time and cost constraints associated with the generation of relevant
-animal data. In this context, alternative approaches to obtain timely
-and fit-for-purpose toxicological information are being developed.
-Amongst others, non-testing, structure-activity based *in silico*
-toxicology methods (also called computational toxicology) are considered
-highly promising. Importantly, they are raising more and more interests
-and getting increased acceptance in various regulatory (e.g. ECHA, 2008;
-EFSA, 2016, 2014; Health Canada, 2016; OECD, 2015) and industrial (e.g.
-Stanton and Kruszewski, 2016; Lo Piparo et al., 2011) frameworks.
+scientific and ethical grounds. In addition, it appears obvious that from
+a resource perspective, the capacity of standard toxicology to address the
+safety of thousands of untested chemicals [@Fowler2011] to which human may be
+exposed is very limited. It has also been recognized that getting rapid insight
+on toxicity of chemicals in case of emergency safety incidents or for early
+prioritization in research and development (safety by design) is a big
+challenge mainly because of the time and cost constraints associated with the
+generation of relevant animal data. In this context, alternative approaches to
+obtain timely and fit-for-purpose toxicological information are being
+developed. Amongst others, non-testing, structure-activity based *in silico*
+toxicology methods (also called computational toxicology) are considered highly
+promising. Importantly, they are raising more and more interests
+and getting increased acceptance in various regulatory (e.g.
+[@ECHA2008, @EFSA2016, @EFSA2014, @HealthCanada2016, @OECD2015]) and industrial (e.g.
+[@Stanton2016, @LoPiparo2011]) frameworks.
 
 For a long time already, computational methods have been an integral
 part of pharmaceutical discovery pipelines, while in chemical food
-safety their actual potentials emerged only recently (Lo Piparo et al.,
-2011). In this later field, an application considered critical is in the
+safety their actual potentials emerged only recently [@LoPiparo2011].
+In this later field, an application considered critical is in the
 establishment of levels of safety concern in order to rapidly and
 efficiently manage toxicologically uncharacterized chemicals identified
 in food. This requires a risk-based approach to benchmark exposure with
-a quantitative value of toxicity relevant for risk assessment (Schilter
-et al., 2014a). Since most of the time chemical food safety deals with
+a quantitative value of toxicity relevant for risk assessment [@Schilter2014].
+Since most of the time chemical food safety deals with
 life-long exposures to relatively low levels of chemicals, and because
 long-term toxicity studies are often the most sensitive in food
 toxicology databases, predicting chronic toxicity is of prime
@@ -128,13 +127,11 @@ The Nestl<U+FFFD><U+FFFD> database can be obtained from the following GitHub lin
 ### Swiss Food Safety and Veterinary Office (FSVO) database
 
 Publicly available data from pesticide evaluations of chronic rat
-toxicity studies from the European Food Safety Authority (EFSA) (EFSA,
-2014), the Joint FAO/WHO Meeting on Pesticide Residues (JMPR) (WHO,
-2011) and the US EPA (US EPA, 2011) were compiled to form the
+toxicity studies from the European Food Safety Authority (EFSA) [@EFSA2014], the Joint FAO/WHO Meeting on Pesticide Residues (JMPR) [@WHO2011] and the US EPA [@EPA2011] were compiled to form the
 FSVO-database. Only studies providing both an experimental NOAEL and an
 experimental LOAEL were included. The LOAELs were taken as they were
 reported in the evaluations. Further details on the database are
-described elsewhere (Zarn et al., 2011; Zarn et al., 2013). The
+described elsewhere [@Zarn2011, @Zarn2013]. The
 FSVO-database consists of 493 rat LOAEL values for 381 unique chemical
 structures. It can be obtained from the following GitHub links:
 
@@ -164,8 +161,8 @@ dataset](https://github.com/opentox/loael-paper/blob/submission/data/test_log10.
 contains data from compounds that occur in both databases. LOAEL values equal
 at five significant digits were considered as duplicates originating from the
 same study/publication and only one instance was kept in the test dataset. The
-test dataset has 375 LOAEL values for `r
-length(unique(t$SMILES))` unique chemical structures and was used for
+test dataset has 375 LOAEL values for 155
+unique chemical structures and was used for
 
 - evaluating experimental variability
 - comparing model predictions with experimental variability.
@@ -401,8 +398,8 @@ same experiments.
 
 
 
-The Nestl<U+FFFD><U+FFFD> database has 567 LOAEL values for `r
-length(levels(m$SMILES))` unique structures, 93 compounds have
+The Nestl<U+FFFD><U+FFFD> database has 567 LOAEL values for
+445 unique structures, 93 compounds have
 multiple measurements with a mean standard deviation (-log10 transformed
 values) of 0.32 (0.56
 mg/kg_bw/day, 0.56 mmol/kg_bw/day)
@@ -439,8 +436,8 @@ datasets. As both datasets contain duplicates medians were used for the
 correlation plot and statistics. It should be kept in mind that the aggregation of duplicated
 measurements into a single median value hides a substantial portion of the
 experimental variability.  Correlation analysis shows a significant (p-value < 2.2e-16)
-correlation between the experimental data in both datasets with r\^2: `r
-round(median.r.square,2)`, RMSE: 0.59
+correlation between the experimental data in both datasets with r\^2:
+0.52, RMSE: 0.59
 
 ![Correlation of median LOAEL values from Nestl<U+FFFD><U+FFFD> and FSVO databases. Data with
   identical values in both databases was removed from
@@ -453,8 +450,8 @@ round(median.r.square,2)`, RMSE: 0.59
 In order to compare the performance of *in silico* read across models with
 experimental variability we are using compounds that occur in both datasets as
 a test set (375 measurements, 155
-compounds). `lazar` read across predictions were obtained for `r
-length(unique(t$SMILES))` compounds, 37
+compounds). `lazar` read across predictions were obtained for
+155 compounds, 37
 predictions failed, because no similar compounds were found in the training
 data (i.e. they were not covered by the applicability domain of the training
 data).
@@ -545,11 +542,11 @@ It is currently acknowledged that there is a strong need for
 toxicological information on the multiple thousands of chemicals to
 which human may be exposed through food. These include for examples many
 chemicals in commerce, which could potentially find their way into food
-(Stanton and Kruszewski, 2016; Fowler et al., 2011), but also substances
-migrating from food contact materials (Grob et al., 2006), chemicals
-generated over food processing (Cottererill et al., 2008), environmental
-contaminants as well as inherent plant toxicants (Schilter et al.,
-2014b). For the vast majority of these chemicals, no toxicological data
+[@Stanton2016, @Fowler2011], but also substances
+migrating from food contact materials [@Grob2006], chemicals
+generated over food processing [@Cotterill2008], environmental
+contaminants as well as inherent plant toxicants [@Schilter2013].
+For the vast majority of these chemicals, no toxicological data
 is available and consequently insight on their potential health risks is
 very difficult to obtain. It is recognized that testing all of them in
 standard animal studies is neither feasible from a resource perspective
@@ -563,7 +560,7 @@ toxicology is thought to play an important role for that.
 In order to establish the level of safety concern of food chemicals
 toxicologically not characterized, a methodology mimicking the process
 of chemical risk assessment, and supported by computational toxicology,
-was proposed (Schilter et al., 2014a). It is based on the calculation of
+was proposed [@Schilter2014]. It is based on the calculation of
 margins of exposure (MoE) between predicted values of toxicity and
 exposure estimates. The level of safety concern of a chemical is then
 determined by the size of the MoE and its suitability to cover the
@@ -578,7 +575,7 @@ carcinogenic potency were developed. In these models, substances in the
 training dataset similar to the query compounds are automatically
 identified and used to derive a quantitative TD50 value. The errors
 observed in these models were within the published estimation of
-experimental variability (Lo Piparo, et al., 2014). In the present
+experimental variability [@LoPiparo2014]. In the present
 study, a similar approach was applied to build models generating
 quantitative predictions of long-term toxicity. Two databases compiling
 chronic oral rat lowest adverse effect levels (LOAEL) as endpoint were
@@ -639,7 +636,7 @@ than chonic studies should be studied. It is likely that more substances
 reflecting a wider chemical domain may be available. To predict such
 shorter duration endpoints would also be valuable for chronic toxicy
 since evidence suggest that exposure duration has little impact on the
-levels of NOAELs/LOAELs (Zarn et al., 2011, 2013).
+levels of NOAELs/LOAELs [@Zarn2011, @Zarn2013].
 
 <!--
 Elena + Benoit
diff --git a/loael.pdf b/loael.pdf
index a9cc71e..e9fb9bf 100644
--- a/loael.pdf
+++ b/loael.pdf
diff --git a/references.bibtex b/references.bibtex
index 8916629..edca4f7 100644
--- a/references.bibtex
+++ b/references.bibtex
@@ -1,132 +1,327 @@
 @Article{Guetlein2012,
-AUTHOR = {Gütlein, Martin and Karwath, Andreas and Kramer, Stefan},
-TITLE = {CheS-Mapper - Chemical Space Mapping and Visualization in 3D},
-JOURNAL = {Journal of Cheminformatics},
-VOLUME = {4},
-YEAR = {2012},
-NUMBER = {1},
-PAGES = {7},
-URL = {http://www.jcheminf.com/content/4/1/7},
-DOI = {10.1186/1758-2946-4-7},
-PubMedID = {22424447},
-ISSN = {1758-2946},
-ABSTRACT = {Analyzing chemical datasets is a challenging task for scientific researchers in the field of chemoinformatics. It is important, yet difficult to understand the relationship between the structure of chemical compounds, their physico-chemical properties, and biological or toxic effects. To that respect, visualization tools can help to better comprehend the underlying correlations. Our recently developed 3D molecular viewer CheS-Mapper (Chemical Space Mapper) divides large datasets into clusters of similar compounds and consequently arranges them in 3D space, such that their spatial proximity reflects their similarity. The user can indirectly determine similarity, by selecting which features to employ in the process. The tool can use and calculate different kind of features, like structural fragments as well as quantitative chemical descriptors. These features can be highlighted within CheS-Mapper, which aids the chemist to better understand patterns and regularities and relate the observations to established scientific knowledge. As a final function, the tool can also be used to select and export specific subsets of a given dataset for further analysis.},
-}
-
-@article{doi:10.1021/ci034207y,
-author = {Andreas Bender and Hamse Y. Mussa, and and Robert C. Glen and Stephan Reiling},
-title = {Molecular Similarity Searching Using Atom Environments, Information-Based Feature Selection, and a Naïve Bayesian Classifier},
-journal = {Journal of Chemical Information and Computer Sciences},
-volume = {44},
-number = {1},
-pages = {170-178},
-year = {2004},
-doi = {10.1021/ci034207y},
-    note ={PMID: 14741025},
-
-URL = { 
-        http://dx.doi.org/10.1021/ci034207y
-    
-},
-eprint = { 
-        http://dx.doi.org/10.1021/ci034207y
-    
-}
-
-}
-
-
-@article{Maunz2013,
-  doi = {10.3389/fphar.2013.00038},
-  url = {http://dx.doi.org/10.3389/fphar.2013.00038},
-  year  = {2013},
-  publisher = {Frontiers Media {SA}},
-  volume = {4},
-  author = {Andreas Maunz and Martin G\"{u}tlein and Micha Rautenberg and David Vorgrimmler and Denis Gebele and Christoph Helma},
-  title = {lazar: a modular predictive toxicology framework},
-  journal = {Frontiers in Pharmacology}
-}
-
-
-
-
-@article{doi:10.1021/ci00057a005,
-author = {David Weininger},
-title = {SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules},
-journal = {Journal of Chemical Information and Computer Sciences},
-volume = {28},
-number = {1},
-pages = {31-36},
-year = {1988},
-doi = {10.1021/ci00057a005},
-
-URL = { 
-        http://dx.doi.org/10.1021/ci00057a005
-    
-},
-eprint = { 
-        http://dx.doi.org/10.1021/ci00057a005
-    
-}
-
-}
-
-@article{OBoyle2011,
-  doi = {10.1186/1758-2946-3-33},
-  url = {http://dx.doi.org/10.1186/1758-2946-3-33},
-  year  = {2011},
-  publisher = {Springer Science and Business Media},
-  volume = {3},
-  number = {1},
-  pages = {33},
-  author = {Noel M OBoyle and Michael Banck and Craig A James and Chris Morley and Tim Vandermeersch and Geoffrey R Hutchison},
-  title = {Open Babel: An open chemical toolbox},
-  journal = {Journal of Cheminformatics}
-}
-
-@article{mazzatorta08,
-author = {Paolo Mazzatorta and Manuel Dominguez Estevez and Myriam Coulet and Benoit Schilter},
-title = {Modeling Oral Rat Chronic Toxicity},
-journal = {Journal of Chemical Information and Modeling},
-volume = {48},
-number = {10},
-pages = {1949-1954},
-year = {2008},
-doi = {10.1021/ci8001974},
-    note ={PMID: 18803370},
-
-URL = { 
-        http://dx.doi.org/10.1021/ci8001974
-    
-},
-eprint = { 
-        http://dx.doi.org/10.1021/ci8001974
-    
+  author =       "Martin Gütlein and Andreas Karwath and Stefan
+                 Kramer",
+  title =        "CheS-Mapper - Chemical Space Mapping and Visualization
+                 in 3D",
+  journal =      "Journal of Cheminformatics",
+  volume =       "4",
+  year =         "2012",
+  number =       "1",
+  pages =        "7",
+  URL =          "http://www.jcheminf.com/content/4/1/7",
+  DOI =          "10.1186/1758-2946-4-7",
+  pubmedid =     "22424447",
+  ISSN =         "1758-2946",
+  abstract =     "Analyzing chemical datasets is a challenging task for
+                 scientific researchers in the field of
+                 chemoinformatics. It is important, yet difficult to
+                 understand the relationship between the structure of
+                 chemical compounds, their physico-chemical properties,
+                 and biological or toxic effects. To that respect,
+                 visualization tools can help to better comprehend the
+                 underlying correlations. Our recently developed 3D
+                 molecular viewer CheS-Mapper (Chemical Space Mapper)
+                 divides large datasets into clusters of similar
+                 compounds and consequently arranges them in 3D space,
+                 such that their spatial proximity reflects their
+                 similarity. The user can indirectly determine
+                 similarity, by selecting which features to employ in
+                 the process. The tool can use and calculate different
+                 kind of features, like structural fragments as well as
+                 quantitative chemical descriptors. These features can
+                 be highlighted within CheS-Mapper, which aids the
+                 chemist to better understand patterns and regularities
+                 and relate the observations to established scientific
+                 knowledge. As a final function, the tool can also be
+                 used to select and export specific subsets of a given
+                 dataset for further analysis.",
 }
 
+@Article{doi:10.1021/ci034207y,
+  author =       "Andreas Bender and  Hamse Y. Mussa and Robert C.
+                 Glen and Stephan Reiling",
+  title =        "Molecular Similarity Searching Using Atom
+                 Environments, Information-Based Feature Selection, and
+                 a Naïve Bayesian Classifier",
+  journal =      "Journal of Chemical Information and Computer
+                 Sciences",
+  volume =       "44",
+  number =       "1",
+  pages =        "170--178",
+  year =         "2004",
+  DOI =          "10.1021/ci034207y",
+  note =         "PMID: 14741025",
+  URL =          "http://dx.doi.org/10.1021/ci034207y",
+  eprint =       "http://dx.doi.org/10.1021/ci034207y",
+}
+
+@Article{Maunz2013,
+  DOI =          "10.3389/fphar.2013.00038",
+  URL =          "http://dx.doi.org/10.3389/fphar.2013.00038",
+  year =         "2013",
+  publisher =    "Frontiers Media {SA}",
+  volume =       "4",
+  author =       "Andreas Maunz and Martin G{\"{u}}tlein and Micha
+                 Rautenberg and David Vorgrimmler and Denis Gebele and
+                 Christoph Helma",
+  title =        "lazar: a modular predictive toxicology framework",
+  journal =      "Frontiers in Pharmacology",
+}
+
+@Article{doi:10.1021/ci00057a005,
+  author =       "David Weininger",
+  title =        "SMILES, a chemical language and information system. 1.
+                 Introduction to methodology and encoding rules",
+  journal =      "Journal of Chemical Information and Computer
+                 Sciences",
+  volume =       "28",
+  number =       "1",
+  pages =        "31--36",
+  year =         "1988",
+  DOI =          "10.1021/ci00057a005",
+  URL =          "http://dx.doi.org/10.1021/ci00057a005",
+  eprint =       "http://dx.doi.org/10.1021/ci00057a005",
+}
+
+@Article{OBoyle2011,
+  DOI =          "10.1186/1758-2946-3-33",
+  URL =          "http://dx.doi.org/10.1186/1758-2946-3-33",
+  year =         "2011",
+  publisher =    "Springer Science and Business Media",
+  volume =       "3",
+  number =       "1",
+  pages =        "33",
+  author =       "Noel M OBoyle and Michael Banck and Craig A James and
+                 Chris Morley and Tim Vandermeersch and Geoffrey R
+                 Hutchison",
+  title =        "Open Babel: An open chemical toolbox",
+  journal =      "Journal of Cheminformatics",
+}
+
+@Article{mazzatorta08,
+  author =       "Paolo Mazzatorta and Manuel Dominguez Estevez and
+                 Myriam Coulet and Benoit Schilter",
+  title =        "Modeling Oral Rat Chronic Toxicity",
+  journal =      "Journal of Chemical Information and Modeling",
+  volume =       "48",
+  number =       "10",
+  pages =        "1949--1954",
+  year =         "2008",
+  DOI =          "10.1021/ci8001974",
+  note =         "PMID: 18803370",
+  URL =          "http://dx.doi.org/10.1021/ci8001974",
+  eprint =       "http://dx.doi.org/10.1021/ci8001974",
 }
 
 @Manual{pls,
-    title = {pls: Partial Least Squares and Principal Component Regression},
-    author = {Bjørn-Helge Mevik and Ron Wehrens and Kristian Hovde Liland},
-    year = {2015},
-    note = {R package version 2.5-0},
-    url = {https://CRAN.R-project.org/package=pls},
-  }
-
-@ARTICLE{Kuhn08,
-    author = {Max Kuhn},
-    title = {Building predictive models in R using the caret package},
-    journal = {J. of Stat. Soft},
-    year = {2008}
-}
-
-@article{Jeliazkova15,
-  author = {Nina Jeliazkova and Charalampos Chomenidis and Philip Doganis and Bengt Fadeel and Roland Grafström and Barry Hardy and Janna Hastings and Markus Hegi and Vedrin Jeliazkov and Nikolay Kochev and Pekka Kohonen and Cristian R. Munteanu and Haralambos Sarimveis and Bart Smeets and Pantelis Sopasakis and Georgia Tsiliki and David Vorgrimmler and Egon Willighagen },
-  title = {The eNanoMapper database for nanomaterial safety information},
-  journal = {Beilstein J. Nanotechnol.},
-  pages = {1609–1634},
-  number = {6},
-  year = {2015},
-  doi = {doi:10.3762/bjnano.6.165}
+  title =        "pls: Partial Least Squares and Principal Component
+                 Regression",
+  author =       "Bjørn-Helge Mevik and Ron Wehrens and Kristian Hovde
+                 Liland",
+  year =         "2015",
+  note =         "R package version 2.5-0",
+  URL =          "https://CRAN.R-project.org/package=pls",
+}
+
+@Article{Kuhn08,
+  author =       "Max Kuhn",
+  title =        "Building predictive models in R using the caret
+                 package",
+  journal =      "J. of Stat. Soft",
+  year =         "2008",
+}
+
+@Article{Jeliazkova15,
+  author =       "Nina Jeliazkova and Charalampos Chomenidis and Philip
+                 Doganis and Bengt Fadeel and Roland Grafström and
+                 Barry Hardy and Janna Hastings and Markus Hegi and
+                 Vedrin Jeliazkov and Nikolay Kochev and Pekka Kohonen
+                 and Cristian R. Munteanu and Haralambos Sarimveis and
+                 Bart Smeets and Pantelis Sopasakis and Georgia Tsiliki
+                 and David Vorgrimmler and Egon Willighagen",
+  title =        "The eNanoMapper database for nanomaterial safety
+                 information",
+  journal =      "Beilstein J. Nanotechnol.",
+  pages =        "1609–1634",
+  number =       "6",
+  year =         "2015",
+  DOI =          "doi:10.3762/bjnano.6.165",
+}
+
+@TechReport{Fowler2011,
+  author =       "B. Fowler and S. Savage and B. Mendez",
+  year =         "2011",
+  title =        "White paper: Protecting public health in the 21st
+                 century: the case for computational toxicology",
+  institution =  "ICF International, Inc.icfi.com.",
+}
+
+@Article{Cotterill2008,
+  author =       "Cotterill, J.V. and Chaudry, M.Q. and Mattews, W. and R.
+                 W. Watkins",
+  year =         "2008",
+  title =        "In silico assessment of toxicity of heat-generated
+                 food contaminants",
+  journal =      "Food Chemical Toxicology",
+  number =       "46(6)",
+  pages =        "1905--1918",
+}
+
+@Article{Grob2006,
+  author =       "Grob, K. and Biedermann, M. and Scherbaum, E. and Roth, M.
+                 and K. Rieger",
+  year =         "2006",
+  title =       "Food contamination with organic materials in
+                 perspective: packaging materials as the largest and
+                 least controlled source? A view focusing on the
+                 European situation",
+  journal =      "Crit. Rev. Food. Sci. Nutr.",
+  number =       "46",
+  pages =        "529--35",
+  DOI =          "10.1080/10408390500295490",
+}
+
+@Article{LoPiparo2011,
+  author =       "Lo Piparo, E. and Worth, A. and Manibusan, A. and Yang, C. and
+                 Schilter, B. and Mazzatorta, P. and Jacobs, M.N. and
+                 Steinkelner, H. and L. Mohimont",
+  year =         "2011",
+  title =        "Use of Computational tools in the field of food
+                 safety",
+  journal =      "Regulatory Toxicology and Pharmacology",
+  number =       "60(3)",
+  pages =        "354--362",
+}
+
+@Article{LoPiparo2014,
+  author =       "Lo Piparo, E. and Maunz, A. and Helma, C. and Vorgrimmler, D. and Schilter, B.",
+  year =         "2014",
+  title =        "Automated and reproducible read-across like models for
+                 predicting carcinogenic potency",
+  journal =      "Regulatory Toxicology and Pharmacology",
+  number =       "70",
+  pages =        "370--378",
+}
+
+@Article{Schilter2014,
+  author =       "Schilter, B. and Benigni, R. and Boobis, A. and Chiodini, A. and
+                 Cockburn, A. and Cronin, M.T. and Lo Piparo, E. and Modi, S. and
+                 Thiel A. and A. Worth",
+  title =        "Establishing the level of safety concern for chemicals
+                 in food without the need for toxicity testing",
+  year =         "2014",
+  journal =      "Regulatory Toxicology and Pharmacology",
+  number =       "68",
+  pages =        "275--298",
+}
+
+@Article{Stanton2016,
+  author =       "Stanton, K. and F. H. Krusezewski",
+  year =         "2016",
+  title =        "Quantifying the benefits of using read-across and in
+                 silico techniques to fullfill hazard data requirements
+                 for chemical categories",
+  journal =      "Regulatory Toxicology and Pharmacology",
+  number =       "81",
+  pages =        "250--259",
+  DOI =          "10.1016/j-yrtph.2016.09.004.",
+}
+
+@Article{Zarn2011,
+  author =       "Zarn, J. A. and B. E. Engeli and J. R. Schlatter",
+  year =         "2011",
+  title =        "Study parameters influencing {NOAEL} and {LOAEL} in
+                 toxicity feeding studies for pesticides: exposure
+                 duration versus dose decrement, dose spacing, group
+                 size and chemical class",
+  journal =      "Regul. Toxicol. Pharmacol.",
+  number =       "61",
+  pages =        "243--250",
+}
+
+@Article{Zarn2013,
+  author =       "Zarn, J. A. and B. E. Engeli and J. R. Schlatter",
+  year =         "2013",
+  title =        "Characterization of the dose decrement in regulatory
+                 rat pesticide toxicity feeding studies",
+  journal =      "Regul. Toxicol. Pharmacol.",
+  number =       "67",
+  pages =        "215--220",
+}
+
+@Article{EFSA2016,
+  author =       {EFSA},
+  year =         "2016",
+  title =        "Guidance on the establishment of the residue
+                  definition for dietary assessment: {EFSA} panel on Plant
+                  Protect Products and their Residues ({PPR})",
+  journal =      "EFSA Journal",
+  number =       "14",
+  pages =        "1--12",
+}
+
+@TechReport{ECHA2008,
+  author =       "{ECHA}",
+  year =         "2008",
+  title =        "Guidance on information requirements and chemical
+                  safety assessment, Chapter R.6: {QSARs} and grouping of
+                 chemicals",
+  institution =  "ECHA",
+}
+
+@Misc{EFSA2014,
+  author =       "{EFSA}",
+  year =         "2014",
+  title =        "Rapporteur Member State assessment reports submitted
+                 for the {EU} peer review of active substances used in
+                 plant protection products",
+  URL =          "http://dar.efsa.europa.eu/dar-web/provision",
+  note =         "accessed 8.1.2015",
+}
+
+@Misc{HealthCanada2016,
+  author =       "{Health Canada}",
+  year =         "2016",
+  URL =          "https://www.canada.ca/en/health-canada/services/chemical-substances/chemicals-management-plan.html",
+}
+
+@InCollection{OECD2015,
+  author =       "{OECD}",
+  year =         "2015",
+  title =        "Fundamental and guiding principles for {(Q)SAR} analysis
+                 of chemicals carcinogens with mechanistic
+                 considerations Monograph 229 {ENV/JM/MONO(2015)46}",
+  chapter =      "229",
+  booktitle =    "Series on Testing and Assessment No 229",
+}
+
+@InCollection{Schilter2013,
+  author =       "Schilter, B. and Constable, A. and Perrin, I.",
+  title =        "Naturally occurring toxicants of plant origin: risk
+                 assessment and management considerations",
+  year =         "2013",
+  booktitle =    "Food Safety Management: a practical guide for
+                 industry",
+  chapter =      "3",
+  editor =       "Y. Motarjemi",
+  publisher =    "Elsevier",
+  pages =        "45--57",
+}
+
+@Misc{EPA2011,
+  title =        "Fact Sheets on New Active Ingredients",
+  author =       "{US EPA}",
+  year =         "2011",
+  note =         "this database was not further maintained by US EPA.
+                 Only data used until 27.11.2014, accessed 27.4.11
+                 A.D.",
+}
+
+@TechReport{WHO2011,
+  author =       "WHO",
+  year =         "2011",
+  title =        "Joint {FAO/WHO} Meeting on Pesticide Residues ({JMPR})
+                 publications",
+  URL =          "http://www.who.int/foodsafety/publications/jmpr-monographs/en/",
+  note =         "accessed 20.3.15 A.D.",
 }