diff options
author | Christoph Helma <helma@in-silico.ch> | 2016-02-01 13:38:45 +0100 |
---|---|---|
committer | Christoph Helma <helma@in-silico.ch> | 2016-02-01 13:38:45 +0100 |
commit | 0f7394eab539df98d350e2a7c76c0b5e6fd0319a (patch) | |
tree | 8d4faa3e049d3b478f5d0bb2fc34d7d97311bf9e | |
parent | 443938e6ab6ac2aaade63bc1c51b7da1ebd960dd (diff) |
workshop presentation improved
-rw-r--r-- | public/enm-workshop.html | 83 | ||||
-rw-r--r-- | public/enm-workshop.rst | 63 |
2 files changed, 89 insertions, 57 deletions
diff --git a/public/enm-workshop.html b/public/enm-workshop.html index f0c6781..05d4308 100644 --- a/public/enm-workshop.html +++ b/public/enm-workshop.html @@ -371,20 +371,20 @@ ul.auto-toc { <div class="slide" id="slide0"> <h1 class="title">Read across toxicity predictions with nano-lazar</h1> -<p class="center">Christoph Helma</p> +<p class="center">Christoph Helma, Denis Gebele, Micha Rautenberg</p> <p class="center">in silico toxicology gmbh</p> <img alt="http://www.enanomapper.net/sites/all/themes/theme807/logo.png" class="center align-center" src="http://www.enanomapper.net/sites/all/themes/theme807/logo.png" /> </div> -<div class="slide" id="requirements"> -<h1>Requirements</h1> -<ul class="simple"> +<div class="slide" id="requirements-for-nanoparticle-read-across"> +<h1>Requirements for nanoparticle read-across</h1> +<ul class="incremental simple"> <li>Nanoparticle characterisation</li> <li>Toxicity measurements</li> </ul> </div> -<div class="slide" id="enanomapper-data-import"> -<h1>eNanoMapper data import</h1> +<div class="slide" id="enanomapper-particle-characterisation"> +<h1>eNanoMapper particle characterisation</h1> <ul class="incremental simple"> <li>Nanoparticles imported: 464</li> <li>Nanoparticles with particle characterisation: 394</li> @@ -403,42 +403,65 @@ ul.auto-toc { <div class="slide" id="selected-data"> <h1>Selected data</h1> <p>Protein corona dataset Au particles (106 particles) -Toxicity endpoint:</p> +Toxicity endpoint: Net cell association (A549 cell line)</p> </div> <div class="slide" id="read-across-procedure"> <h1>Read across procedure</h1> <ul class="incremental simple"> -<li>Identify relevant fragments (significant correlation with toxicity) -TODO list of fragments, number</li> -<li>Calculate similarities (weighted cosine similarity, correlation coefficients = weights)</li> -<li>Identify neighbors (particles with more than 0.95 similarity)</li> -<li>Calculate prediction (weighted average from neighbors, similarities = weights)</li> +<li>Identify relevant properties (statistically significant correlation with toxicity: 14 from 30 properties)</li> +<li>Calculate similarities (weighted cosine similarity with correlation coefficients as weights)</li> +<li>Identify neighbors (particles with similarity > 0.95)</li> +<li>Calculate prediction (weighted average from neighbors with similarities as weights)</li> </ul> +<p class="incremental">Algorithms for feature selection, similarity calculation and predictions may change in the future.</p> </div> -<div class="slide" id="future-development"> -<h1>Future development</h1> +<div class="slide" id="future-development-i"> +<h1>Future development (I)</h1> <ul class="simple"> <li>Validation of predictions</li> <li>Applicability domain/reliability of predictions</li> -<li>Accuracy improvements: -- additional data -- feature selection -- similarity calculation -- predictions (local regression models)</li> -<li>Usability improvements: -- additional data (extension of applicability domain, additional endpoints and chemistries) -- inclusion of ontologies -- descriptor calculation directly from core and coating chemistries</li> +<li>Accuracy improvements:<ul> +<li>additional data</li> +<li>feature selection</li> +<li>similarity calculation</li> +<li>predictions (local regression models)</li> +</ul> +</li> </ul> </div> -<div class="slide" id="webinterface"> -<h1>Webinterface</h1> -<p><a class="reference external" href="https://nano-lazar.in-silico.ch/predict">https://nano-lazar.in-silico.ch/predict</a></p> -<p>Your recommendations?</p> +<div class="slide" id="id1"> +<h1>Future development (I)</h1> +<ul class="simple"> +<li>Usability improvements:<ul> +<li>additional data (extension of applicability domain, additional endpoints and chemistries)</li> +<li>inclusion of ontologies</li> +<li>inclusion of protein corona characterisation?</li> +<li>particle characterisation without experimental data<ul> +<li>descriptor calculation from core and coating chemistries</li> +<li>ontological descriptors</li> +</ul> +</li> +</ul> +</li> +</ul> </div> -<div class="slide" id="source-code"> -<h1>Source code</h1> -<p><a class="reference external" href="https://github.com/opentox/nano-lazar">https://github.com/opentox/nano-lazar</a></p> +<div class="slide" id="nano-lazar"> +<h1>nano-lazar</h1> +<table class="docutils field-list" frame="void" rules="none"> +<col class="field-name" /> +<col class="field-body" /> +<tbody valign="top"> +<tr class="field"><th class="field-name">Webinterface:</th><td class="field-body"><a class="reference external" href="https://nano-lazar.in-silico.ch/predict">https://nano-lazar.in-silico.ch/predict</a></td> +</tr> +<tr class="field"><th class="field-name">Presentation:</th><td class="field-body"><a class="reference external" href="https://nano-lazar.in-silico.ch/predict/enm-workshop.html">https://nano-lazar.in-silico.ch/predict/enm-workshop.html</a></td> +</tr> +<tr class="field"><th class="field-name">Source code:</th><td class="field-body"><a class="reference external" href="https://github.com/enanomapper/nano-lazar-gui">https://github.com/enanomapper/nano-lazar-gui</a></td> +</tr> +<tr class="field"><th class="field-name">Issues:</th><td class="field-body"><a class="reference external" href="https://github.com/enanomapper/nano-lazar-gui/issues">https://github.com/enanomapper/nano-lazar-gui/issues</a></td> +</tr> +</tbody> +</table> +<p>Your comments, ideas, recommendations?</p> </div> </div> </body> diff --git a/public/enm-workshop.rst b/public/enm-workshop.rst index 26a3524..d54f4cb 100644 --- a/public/enm-workshop.rst +++ b/public/enm-workshop.rst @@ -1,29 +1,29 @@ .. |date| date:: - ============================================================= Read across toxicity predictions with nano-lazar ============================================================= .. class:: center - Christoph Helma + Christoph Helma, Denis Gebele, Micha Rautenberg in silico toxicology gmbh .. image:: http://www.enanomapper.net/sites/all/themes/theme807/logo.png :align: center -Requirements -============ +Requirements for nanoparticle read-across +========================================= -- Nanoparticle characterisation -- Toxicity measurements +.. class:: incremental + - Nanoparticle characterisation + - Toxicity measurements -eNanoMapper data import -======================= +eNanoMapper particle characterisation +===================================== .. class:: incremental @@ -38,52 +38,61 @@ eNanoMapper toxicity endpoints .. class:: incremental -- Toxicity endpoints: 41 -- Toxicity endpoints with more than one measurement value: 22 -- Toxicity endpoints with more than 10 measurements: 2 + - Toxicity endpoints: 41 + - Toxicity endpoints with more than one measurement value: 22 + - Toxicity endpoints with more than 10 measurements: 2 Selected data ============= Protein corona dataset Au particles (106 particles) -Toxicity endpoint: +Toxicity endpoint: Net cell association (A549 cell line) Read across procedure ===================== .. class:: incremental -- Identify relevant fragments (significant correlation with toxicity) - TODO list of fragments, number -- Calculate similarities (weighted cosine similarity, correlation coefficients = weights) -- Identify neighbors (particles with more than 0.95 similarity) -- Calculate prediction (weighted average from neighbors, similarities = weights) + - Identify relevant properties (statistically significant correlation with toxicity: 14 from 30 properties) + - Calculate similarities (weighted cosine similarity with correlation coefficients as weights) + - Identify neighbors (particles with similarity > 0.95) + - Calculate prediction (weighted average from neighbors with similarities as weights) -Future development -================== + Algorithms for feature selection, similarity calculation and predictions may change in the future. + +Future development (I) +====================== - Validation of predictions - Applicability domain/reliability of predictions - Accuracy improvements: + - additional data - feature selection - similarity calculation - predictions (local regression models) +Future development (I) +====================== + - Usability improvements: + - additional data (extension of applicability domain, additional endpoints and chemistries) - inclusion of ontologies - - descriptor calculation directly from core and coating chemistries + - inclusion of protein corona characterisation? + - particle characterisation without experimental data -Webinterface -============ + - descriptor calculation from core and coating chemistries + - ontological descriptors -https://nano-lazar.in-silico.ch/predict +nano-lazar +===================== -Your recommendations? +:Webinterface: https://nano-lazar.in-silico.ch/predict +:Presentation: https://nano-lazar.in-silico.ch/predict/enm-workshop.html +:Source code: https://github.com/enanomapper/nano-lazar-gui +:Issues: https://github.com/enanomapper/nano-lazar-gui/issues -Source code -=========== +Your comments, ideas, recommendations? -https://github.com/opentox/nano-lazar |