Skip to content

Commit

Permalink
mpc file
Browse files Browse the repository at this point in the history
  • Loading branch information
@amva13
amva13 authored Nov 12, 2024
1 parent 32d3be7 commit 24b2d26
Showing 1 changed file with 365 additions and 0 deletions.
365 changes: 365 additions & 0 deletions templates/single_pred_tasks/index(1).html
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,365 @@

<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Molecular Property Cliff (a.k.a. Activity Cliff) Prediction - TDC</title>
<link rel="stylesheet" href="/assets/css/app.css">
<link rel="shortcut icon" type="image/png"
href="/favicon.png"
/>
<script defer src="https://use.fontawesome.com/releases/v5.3.1/js/all.js"></script>
<!-- Begin Jekyll SEO tag v2.8.0 -->
<title>Molecular Property Cliff (a.k.a. Activity Cliff) Prediction | TDC</title>
<meta name="generator" content="Jekyll v3.9.3" />
<meta property="og:title" content="Molecular Property Cliff Prediction" />
<meta property="og:locale" content="en_US" />
<meta name="description" content="Artificial intelligence foundation for therapeutic science" />
<meta property="og:description" content="Artificial intelligence foundation for therapeutic science" />
<link rel="canonical" href="http://localhost:4000/single_pred_tasks/MPC/" />
<meta property="og:url" content="http://localhost:4000/single_pred_tasks/MPC/" />
<meta property="og:site_name" content="TDC" />
<meta property="og:type" content="website" />
<meta name="twitter:card" content="summary" />
<meta property="twitter:title" content="Molecular Property Cliff Prediction" />
<script type="application/ld+json">
{"@context":"https://schema.org","@type":"WebPage","description":"Artificial intelligence foundation for therapeutic science","headline":"Molecular Property Cliff Prediction","url":"http://localhost:4000/single_pred_tasks/MPC/"}</script>
<!-- End Jekyll SEO tag -->
<script async src="https://www.googletagmanager.com/gtag/js?id=UA-183522810-1"></script>
<script>
window['ga-disable-UA-183522810-1'] = window.doNotTrack === "1" || navigator.doNotTrack === "1" || navigator.doNotTrack === "yes" || navigator.msDoNotTrack === "1";
window.dataLayer = window.dataLayer || [];
function gtag(){dataLayer.push(arguments);}
gtag('js', new Date());
gtag('config', 'UA-183522810-1');
</script><!-- head scripts --></head>

<body>

<nav class="navbar is-primary" >
<div class="container">
<div class="navbar-brand">
<a class="navbar-brand" href="/">
<span><img src="/logonav.png" alt="Logo" style="height: auto; width: auto; max-height: 45px; max-width: 250px;"></span>
</a>
<a role="button" class="navbar-burger burger" aria-label="menu" aria-expanded="false" data-target="navMenu">
<span aria-hidden="true"></span>
<span aria-hidden="true"></span>
<span aria-hidden="true"></span>
</a>
</div>
<div class="navbar-menu" id="navMenu">
<div class="navbar-end">



<a href="/" class="navbar-item ">Home</a>



<a href="/start/" class="navbar-item ">Start</a>



<div class="navbar-item has-dropdown is-hoverable">
<a href="/overview/" class="navbar-link ">Datasets</a>
<div class="navbar-dropdown">

<a href="/overview/" class="navbar-item ">Overview</a>

<a href="/single_pred_tasks/overview" class="navbar-item ">Single-instance Prediction</a>

<a href="/multi_pred_tasks/overview" class="navbar-item ">Multi-instance Prediction</a>

<a href="/generation_tasks/overview" class="navbar-item ">Generation</a>

</div>
</div>



<div class="navbar-item has-dropdown is-hoverable">
<a href="/fct_overview/" class="navbar-link ">Data Functions</a>
<div class="navbar-dropdown">

<a href="/fct_overview/" class="navbar-item ">Overview</a>

<a href="/functions/data_evaluation/" class="navbar-item ">Model Evaluation</a>

<a href="/functions/data_split/" class="navbar-item ">Dataset Splits</a>

<a href="/functions/data_process/" class="navbar-item ">Data Processing</a>

<a href="/functions/oracles/" class="navbar-item ">Molecule Generation Oracles</a>

</div>
</div>



<div class="navbar-item has-dropdown is-hoverable">
<a href="/benchmark/overview/" class="navbar-link ">Leaderboards</a>
<div class="navbar-dropdown">

<a href="/benchmark/overview/" class="navbar-item ">Guidelines</a>

<a href="/benchmark/admet_group/overview" class="navbar-item ">ADMET Group</a>

<a href="/benchmark/drugcombo_group/overview" class="navbar-item ">DrugCombo Group</a>

<a href="/benchmark/docking_group/overview" class="navbar-item ">Docking Group</a>

<a href="/benchmark/dti_dg_group/overview" class="navbar-item ">DTI DG Group</a>

<a href="/benchmark/scdti_group/overview" class="navbar-item ">Single-cell DTI Group</a>

<a href="/benchmark/proteinpeptide_group/overview" class="navbar-item ">Protein-Peptide Binding Affinity Group</a>

<a href="/benchmark/counterfactual_group/overview" class="navbar-item ">Counterfactual Prediction Group</a>

<a href="/benchmark/clinical_trial/overview" class="navbar-item ">Clinical Trial Outcome Prediction Group</a>

</div>
</div>



<a href="https://huggingface.co/tdc" class="navbar-item ">HF Models</a>



<a href="/news/" class="navbar-item ">News</a>



<a href="/team/" class="navbar-item ">Team</a>


<a href="https://arxiv.org/abs/2102.09548" class="navbar-item">NeurIPS Paper</a>
<a href="https://www.nature.com/articles/s41589-022-01131-2" class="navbar-item">Nat Chem Bio Paper</a>
<a href="https://tdc.readthedocs.io" class="navbar-item">Docs</a>
<a href="https://github.com/mims-harvard/TDC" class="navbar-item">GitHub</a>

</div>

</div>
</div>
</nav>





<section class="section">
<div class="container">
<div class="columns">

<div class="column is-4-desktop is-4-tablet">


<aside class="menu">

<p class="menu-label"></p>
<ul class="menu-list">

<li>
<a href="/single_pred_tasks/overview/" class=""><strong>Single-instance Prediction Problem</strong></a>

<ul>


<li><a href="/single_pred_tasks/adme/" class=""><code>ADME</code> Pharmaco-kinetics</a></li>



<li><a href="/single_pred_tasks/tox/" class=""><code>Tox</code> Toxicity</a></li>



<li><a href="/single_pred_tasks/hts/" class=""><code>HTS</code> High-Throughput Screening</a></li>



<li><a href="/single_pred_tasks/qm/" class=""><code>QM</code> Quantum Mechanics</a></li>



<li><a href="/single_pred_tasks/yields/" class=""><code>Yields</code> Reaction Yields Prediction</a></li>



<li><a href="/single_pred_tasks/epitope/" class=""><code>Epitope</code> Epitope Prediction</a></li>



<li><a href="/single_pred_tasks/develop/" class=""><code>Develop</code> Developability Prediction</a></li>



<li><a href="/single_pred_tasks/CRISPROutcome/" class="is-active"><code>CRISPROutcome</code> CRISPR Repair Prediction</a></li>


<li><a href="/single_pred_tasks/MPC/" class="is-active"><code>MPC</code> Molecular Property Cliff Prediction</a></li>



</ul>

</li>

</ul>

<p class="menu-label"></p>
<ul class="menu-list">

<li>
<a href="/overview/" class=""><strong>Others</strong></a>

<ul>


<li><a href="/multi_pred_tasks/overview/" class="">Multi-instance Prediction Problem</a></li>



<li><a href="/generation_tasks/overview/" class="">Generation Problem</a></li>


</ul>

</li>

</ul>

</aside>
</div>

<div class="column is-8">







<div class="contents">
<div class="menu">
<p class="menu-label">Dataset Index</p>
<ul class="menu-list">
<li><a href="#leenay-et-al">Wan Xiang et al.</a></li>
</ul>
</div>
</div>




<div class="content">
<h1 id="crispr-repair-outcome-prediction-task-overview">Molecular Property Cliff Prediction Task Overview</h1>

<div class="box">


<p class="is-size-6"> <strong> Definition: </strong>
Activity cliffs are molecules with small differences in structure but large differences in potency. Activity cliffs play an important role in drug discovery, but the bioactivity of activity cliff compounds are notoriously difficult to predict.
</p>

<p class="is-size-6"> <strong> Impact: </strong>
Predicting molecular activity and modeling quantitative structure-activity relationships are crucial for drug discovery. Graph neural networks use molecular structures as frameworks to evaluate the biological activity of chemical compounds. They guide the selection and optimization of candidates for further development. However, current models often overlook activity cliffs (ACs), where structurally similar molecules exhibit different bioactivities. This oversight is due to latent spaces primarily optimized for structural features (Wan Xiang et al.).
</p>

<p class="is-size-6"> <strong> Generalization: </strong>
ACs, or activity cliffs, occur when structurally similar molecules have very different biological activities, creating challenges for accurate modeling. This is especially problematic in Graph Neural Networks (GNNs), where similar molecules are closely grouped in the latent space, leading to inaccurate predictions when their activities are significantly different. Overcoming these challenges is essential for improving the accuracy and reliability of forecasts related to molecular activities. Dealing with ACs directly at the level of compound pairs by predicting whether a matched molecular pair (MMP) forms an AC based on a predefined activity threshold can be a more practical approach. For example, they classify it as an MMP-cliff if the activity difference is more significant than 100-fold or an MMP-nonCliff if the activity difference is less than 10-fold. It has been observed that QSAR regression models have low sensitivity to ACs when the activities of both compounds in the MMP are unknown, as they are absent from the training set. Developing techniques to improve AC sensitivity could enhance the performance of QSAR models and offer a promising direction for future research (Wan Xiang et al.)
</p>

<p class="is-size-6"> <strong> Product: </strong> Small Molecule</p>

<p class="is-size-6"> <strong> Pipeline: </strong> Hit identification and hit-to-lead optimization </p>

</div>

<h3 id="leenay-et-al">Wan Xiang et al.</h3>

<p class="is-size-6"> <strong> Dataset Description: </strong> Benchmark datasets of molecular property cliff (MPC) in ACANet paper. Includes: 1) The 9 datasets of low sample size and narrow scaffolds (LSSNS) for molecular activity prediction LSSNS, 2) The 30 datasets of high sample size and mixed scaffolds (HSSMS) for molecular activity prediction. Datasets are the molecular activity prediction benchmark datasets that from MoleculeACE, 3) The 3 matched molecular pair (MMP) datasets of activity cliff classification. Datasets are from ACGCN, 4) The 10 datasets of ADMET properties in delta prediction. Datasets are from DeepDelta. More information can be found https://github.com/bidd-group/MPCD . </p>

<p class="is-size-6"> Information on each individual dataset can be found https://github.com/bidd-group/MPCD . </p>

<p class="is-size-6"> <strong> Task Description: </strong>Regression. Given a SMILES sequence, predict the activity cliff of the small molecule compound. </p>

<p class="is-size-6"> <strong> Dataset Statistics: </strong> More information on each individual dataset can be found https://github.com/bidd-group/MPCD .</p>

<p class="is-size-6"> <strong> Dataset Split: </strong> <span class="tag is-info is-light">Random Split</span> <span class="tag is-info is-light">Scaffold Split</span> </p>

<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="kn">from</span> <span class="nn">tdc.single_pred</span> <span class="kn">import</span> <span class="n">MPC</span>
<span class="n">data</span> <span class="o">=</span> <span class="n">MPC</span><span class="p">(</span><span class="n">name</span> <span class="o">=</span> <span class="s">"INSERT_URL_HERE"</span><span class="n">)</span> # url from the source github repo https://github.com/bidd-group/MPCD/tree/main/dataset
<span class="n"># example url: https://raw.githubusercontent.com/bidd-group/MPCD/main/dataset/ADMET/DeepDelta_benchmark/Caco2.csv </span>
<span class="n">split</span> <span class="o">=</span> <span class="n">data</span><span class="p">.</span><span class="n">get_data</span><span class="p">()</span>
</code></pre></div></div>

<p class="is-size-6"> We additionally support direct retrieval from the MoleculeACE API [2] for those datasets. You can call: </p>

<div class="language-python highlighter-rouge"><div class="highlight"><pre class="highlight"><code>
<span class="n">data</span> <span class="o">=</span> <span class="n">MPC</span><span class="p">(</span><span class="n">name</span> <span class="o">=</span> <span class="s">"INSERT_MOLECULEACE_HERE"</span>, <span class="n">get_from_gh</span> <span class="o">=</span> <span class="n">False</span><span class="n">)</span> # name from MoleculeACE API https://github.com/molML/MoleculeACE?tab=readme-ov-file
</code></pre></div></div>


<p class="is-size-6"> <strong> References: </strong> </p>

<p><a href="https://doi.org/10.21203/rs.3.rs-2988283/v1">[1] Wan Xiang, et al. “Online triplet contrastive learning enables efficient cliff awareness in molecular activity prediction” 28 June 2023, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-2988283/v1].</a></p>
<p><a href="https://github.com/molML/MoleculeACE?tab=readme-ov-file">[2] van Tilborg et al. "Exposing the Limitations of Molecular Machine Learning with Activity Cliffs.", Journal of Chemical Information and Modeling, 2022, 62 (23), 5938-5951. DOI: 10.1021/acs.jcim.2c01073.</a></p>

<p class="is-size-6"> <strong> Dataset License: </strong><a href="https://creativecommons.org/licenses/by/4.0/">CC BY 4.0</a>.</p>

<hr />


</div>
</div>

</div>
</div>
</section>

<footer class="footer">
<div class="container">

<div class="columns is-mobile">
<div class="column is-8 has-text-left is-vcentered">
<a class="navbar-brand" href="/">
<span><img src="/tdc_horizontal.png" alt="Logo" style="max-height: 40px; max-width: 250px;"></span>
</a>
</div>
<div class="column is-4 has-text-right is-vcentered">
<a href="https://arxiv.org/abs/2102.09548">
<span class="icon is-large">
<i class="fas fa-file-alt fa-3x"></i>
</span>
</a>

<a href="https://github.com/mims-harvard/TDC">
<span class="icon is-large">
<i class="fas fab fa-github fa-3x"></i>
</span>
</a>

<a href="https://twitter.com/ProjectTDC">
<span class="icon is-large">
<i class="fas fab fa-twitter fa-3x"></i>
</span>
</a>

<a href="https://join.slack.com/t/pytdc/shared_invite/zt-x0ujg5v6-zwtQZt83fhRdgrYjXRFz5g">
<span class="icon is-large">
<i class="fas fab fa-slack fa-3x"></i>
</span>
</a>
</div>
</div>

</div>
</footer>

<script src="/assets/js/app.js" type="text/javascript"></script><!-- footer scripts --></body>
</html>

0 comments on commit 24b2d26

Please sign in to comment.