COMIC: Explainable Drug Repurposing via Contrastive Masking for Interpretable Connections

COMIC: Explainable Drug Repurposing via Contrastive Masking for Interpretable Connections | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results COMIC: Explainable Drug Repurposing via Contrastive Masking for Interpretable Connections Naafey Aamer , Muhammad Nabeel Asim , Andreas Dengel doi: https://doi.org/10.1101/2025.03.26.645428 Naafey Aamer 1 Department of Computer Science, Rhineland-Palatinate Technical University of Kaiserslautern-Landau , Kaiserslautern, 67663, Germany . Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: naafey.aamer{at}dfki.de Nabeel.Asim{at}dfki.de Muhammad Nabeel Asim 2 German Research Center for Artificial Intelligence GmbH , Kaiserslautern, 67663, Germany . 3 Intelligentx GmbH (intelligentx.com), Kaiserslautern, Germany . Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: naafey.aamer{at}dfki.de Nabeel.Asim{at}dfki.de Andreas Dengel 1 Department of Computer Science, Rhineland-Palatinate Technical University of Kaiserslautern-Landau , Kaiserslautern, 67663, Germany . 2 German Research Center for Artificial Intelligence GmbH , Kaiserslautern, 67663, Germany . Find this author on Google Scholar Find this author on PubMed Search for this author on this site Abstract Full Text Info/History Metrics Supplementary material Data/Code Preview PDF Abstract Many diseases worldwide remain untreated due to the slow and expensive process of drug development. Repurposing existing FDA-approved drugs offers a faster solution, especially with the assistance of artificial intelligence. Despite advancements in AI-driven drug repurposing, current approaches either have lackluster performance or fail to highlight the intricate pathways through which drugs act on diseases. The clinical utility of AI-driven drug repurposing remains constrained by these limitations, particularly for rare and undertreated diseases where data is scarce. To address the need for a precise and explainable predictor, this paper introduces COMIC ( CO ntrastive M asking with I nterpretable C onnections), a predictor that employs a multi channel architecture consisting of a feature masking branch, which identifies critical drug-disease interaction patterns by extracting the most informative features, and a path masking branch, which highlights relevant biological pathways through which drugs exert their therapeutic effects. Comprehensive evaluation of the COMIC predictor on the PrimeKG knowledge graph (comprising 17,080 diseases, and 4M+ relationships) with nine distinct disease area splits demonstrated a 9.55% average performance improvement over the current state-of-the-art. The practical applicability of the proposed predictor is evaluated on a set of the most recent 30 FDA-approved repurposed drug disease pairs. The COMIC predictor successfully identified 21 of these pairs with high confidence scores. To facilitate real-time drug repurposing investigations, we have developed a publicly available web-based interface for the COMIC predictor ( https://drp.opendfki.de/ ). This application takes disease names as input and returns a ranked list of potential repurposing candidates, along with predicted mechanistic pathways elucidating the drug-disease interactions. 1 Introduction The drug repurposing paradigm encompasses interactions of drugs and disease from two perspectives: indications and contraindications. Indications refer to the FDA approved medical uses of a drug, specifically the diseases the drug is approved to treat [ 1 ]. These approvals are based on exhaustive clinical research, including clinical trials, real-world evidence, and comprehensive safety and efficacy assessments. Predicting indications is the most significant step for drug repurposing as it encapsulates the potential of a drug to treat a specific disease, forming the primary objective. On the other hand, Contraindications are situations where a drug should not be used because it may cause harm to patients with certain diseases [ 1 ]. In drug repurposing, understanding contraindications is a secondary step to ensure that a drug being considered for a new use does not pose risks to individuals with specific health conditions. But at the same time, uncovering new indications remains the key objective. The motivation for finding new indications is grounded in the understanding that drugs often have cascading effects, meaning they can influence multiple biological pathways beyond their primary targets ([ 2 ]). Around 30% of FDA-approved drugs have received approval for additional indications beyond their initial intended use with some accumulating over ten new therapeutic applications ([ 3 ]). However, many successful drug repurposing efforts have historically resulted from serendipitous discoveries ([ 4 ]). These include cases where clinicians observed unexpected benefits when drugs were used for conditions other than their original purpose, such as the use of Metformin to treat polycystic ovary syndrome and the unexpected drug benefits in off-label settings of Thalidomide for multiple myeloma ([ 5 ]) ([ 6 ]). In contrast to random discoveries, a systematic strategy could enable the repurposing of existing drugs for new therapeutic uses with fewer side effects ([ 7 ]). For instance, systematic analysis of drug-protein interaction networks led to the identification of kinase inhibitors like Baricitinib for COVID-19 treatment by targeting both viral entry and inflammation pathways, while network-based screening revealed the potential of antihypertensive drugs like Prazosin for treating post-traumatic stress disorder (PTSD) through shared molecular pathways ([ 8 ])([ 9 ]). One of these strategies is the potential matching of drugs to new indications through the analysis of medical knowledge graphs (KGs) ([ 10 ]). By evaluating the actions of drugs on cell signaling pathways, gene expression profiles, and disease phenotypes, potential therapeutic candidates can be identified ([ 2 ]). The underlying patterns of these actions are learned by geometric deep-learning models, which are trained on extensive medical KGs ([ 11 ]). These optimized models can efficiently align disease signatures with drug candidates by analyzing their biological networks ([ 11 ]). While approaches that utilize knowledge graphs (KGs) and deep learning models have improved the systematic identification of repurposing candidates, most current approaches function as ‘black boxes.’ These comprehensive, state-of-the-art predictors [ 12 ] provide predictions without offering clear, human-understandable explanations of their underlying reasoning ([ 13 ]). This lack of explainability poses a significant barrier to clinical adoption and validation of computational drug repurposing predictions ([ 14 ]). From a biological perspective, the identification of specific pathways connecting drug candidates to disease targets is crucial for several reasons. Biological pathways represent the fundamental mechanisms through which drugs exert their therapeutic effects. Through explainable methods, accurate prediction of pathways provides mechanistic validation that a predicted drug-disease association operates through biologically plausible routes, distinguishing meaningful therapeutic opportunities from spurious correlations ([ 15 ]). Moreover, pathway-level explanations enable researchers to assess safety implications by identifying potential off-target effects or conflicting regulatory mechanisms that might contraindicate use in specific patient populations ([ 16 ]) ([ 17 ]). Additionally, a drug repurposed for a disease can be unrelated to the indication it was initially studied for, as exemplified by Metformin’s evolution from a diabetes treatment to potential anti-cancer properties ([ 18 ]). Thus, mechanistic explanations for each prediction are crucial for building clinical trust in AI. On the other hand, existing predictors that leverage knowledge graphs to predict drug-disease interactions, and give pathway level interpretability, struggle to balance predictive power with plausible explanations. Some of these approaches rely on predefined path schemas [ 19 ], limiting their ability to discover novel biological mechanisms. Others use complex learning strategies (such as reinforcement learning) [ 20 ] [ 21 ] that require extensive tuning, making them less adaptable across different diseases. Rulebased methods provide clear explanations [ 22 ] but struggle with generalization, while deep learning based predictors often prioritize accuracy at the cost of transparency [ 23 ]. These challenges highlight the need for approaches that can provide biologically meaningful explanations without sacrificing predictive performance. Apart from the trade-off between performance and explainability, another major challenge that predictors face is data scarcity. Around 92% of 17,080 examined diseases have no approved indications and this scarcity is even more pronounced for rare diseases, where approximately 95% lack FDA-approved treatments and up to 85% have no promising drug candidates ([ 7 ], [ 24 ]). This limited data constrains predictive methods, leaving them with too few examples to learn meaningful drug-disease relationships. These minimal therapeutic options and limited molecular understanding poses a significant challenge for drug-repurposing predictors ([ 7 ]). These challenges highlight the need for predictors that can effectively learn from sparse data while providing mechanistic insights to validate repurposing predictions, particularly when exploring novel therapeutic applications. To address this gap, we introduce COMIC, a predictor that utilizes CO ntrastive M asking with I nterpretable C onnections and aims to balances predictive performance with interpretability. The primary motivation behind incorporating contrastive masking is to identify and preserve the most informative features within sparse data. Through contrastive learning, these networks distinguish meaningful patterns and paths, allowing the predictor to capture both common and rare but significant interaction signatures. This selective approach makes the prediction process interpretable at multiple levels: from identifying critical interaction features through the feature selection network to discovering relevant biological pathways through the path selection network. This enables tracking how each prediction emerges through selected features and pathways while maintaining strong predictive performance. The proposed predictor is evaluated across nine disease categories of the PrimeKG knowledge graph (comprising 17,080 diseases, and 4M+ relationships). COMIC demonstrated a significant performance boost over the current state-of-the-art [ 7 ], achieving an average improvement of 4.29% in indication prediction and 14.82% in contraindication prediction, resulting in an overall average gain of 9.55% across both interaction types. To validate COMIC’s real-world applicability, we evaluated its performance on 30 FDA-approved (2022-2025) repurposed indications that are not present in PrimeKG. The predictor successfully predicted 21 out of 30 (70%) approved indications, with 10 predictions ranking above the 90th percentile. Compared to the state-of-the-art [ 7 ], COMIC showed superior discrimination ability. These results demonstrate COMIC’s practical utility while highlighting opportunities for improvement in challenging domains like oncology and neurodegenerative diseases. To facilitate broad adoption, we have deployed our predictor as a web application that provides both predictions and their comprehensive explanations for indications, contraindications, and off-label use https://drp.opendfki.de/ . 2 Related Work Recent technological advances have led to the rapid development of numerous knowledge graph-based drug repurposing predictors. These predictors can be categorized into two distinct groups: (1) predictors that solely identify interactions between drugs and diseases and (2) predictors that not only make interaction predictions but also provide explanations. As discussed in the previous section, explanations play a crucial role in drug repurposing; therefore, this section only provides a high-level overview of the second category of predictors. These explainable predictors have evolved into distinct streams, each employs different strategies for model interpretability, ranging from attention-based path weighting to rule-based reasoning systems. While these methods have made significant progress in explaining drug-disease associations, they involve trade-offs between interpretability, computational efficiency, and biological plausibility. KGML-xDTD employs a hybrid approach combining knowledge graph embeddings with reinforcement learning to identify drug-disease associations ([ 19 ]). MIN-ERVA and PoLo employ similar reinforcement learning agents to traverse KGs, generating rule-guided paths that mimic human hypothesis generation ([ 21 ]) ([ 20 ]). While KMGL-xDTD’s attention-weighted metapaths capture biological relationships, it is constrained by predefined metapath schemas and reinforcement learning based predictors require extensive reward engineering and may overfit to specific traversal patterns, especially on large graphs like PrimeKG. XG4Repo employs the RNNLogic+’s framework which uses neural logical reasoning in which a RNN generates logic rules for drug repositioning patterns ([ 25 ]) ([ 26 ]). Hybrid frameworks like SAFRAN integrate KG embeddings with rule-based systems, using Noisy-OR functions to aggregate metapath scores ([ 22 ]). These rules based system are often constrained by the expressiveness of their generated rules, which may fail to generalize across diverse drug-disease relationships. A promising approach involves using graph explanation models to interpret Graph Neural Networks (GNNs). These methods analyze GNNs by selectively masking parts of the graph and assessing how the absence of these parts impacts predictions, thereby highlighting the most influential components in the decision-making process. RD-Explainer adapts GNNExplainer to generate semantic subgraphs highlighting genes, symptoms, and pathways ([ 27 ]) ([ 28 ]). TxGNN integrates message-passing graph neural networks with metric learning to rank drug-disease pairs across 17,080 diseases ([ 7 ]), achieving state-of-the-art performance. However, its interpretability relies on a post-hoc GraphMask-based explainer to identify critical pathways ([ 29 ]). COMIC builds upon TxGNN’s extensive disease coverage and GNN architecture, introducing a contrastive masking-based predictor that provides end-to-end interpretability. Unlike TxGNN, COMIC learns to distinguish relevant features and pathways during training, making interpretability an inherent part of its process. This design allows COMIC to jointly optimize for both prediction accuracy and explanation quality, adapting to complex feature interactions while maintaining strong performance across the full disease spectrum. 3 Materials and Methods This section provides a detailed overview of the COMIC predictor’s architecture, describes the dataset used for its evaluation, and outlines the metrics employed to assess both its predictive performance and the interpretability of its explanations. 3.1 The Proposed Predictor As illustrated in Figure 1 , COMIC takes a knowledge graph G as input, which is formally represented as a collection of triples, where each triple t is defined as: Download figure Open in new tab Fig. 1 Overview of the COMIC architecture, illustrating the constrastive masking approach where h and l denote the head and tail entities respectively, and r represents the relation between them. The first step is embedding generation. For each entity and relation, dense vector representations are generated: These embeddings are generated through Xavier initialization and are normalized for stability [ 30 ]. The normalized entity and relation embeddings then flow into a multi-channel architecture which has two branches: the feature masking branch and the path masking branch. 3.1.1 Feature Masking Branch This branch first processes each input triple through the Triple Encoding layer and its output is fed into the Feature Masking layer. The triple encoding layer first concatenates the head, relation, and tail embeddings to preserve their sequential structure. Positional encoding is then applied to enhance this sequence with ordering information. The position-enhanced sequence is passed through a self-attention layer, enabling the triple embedding to maintain the sequence of the triple elements. The attended representation undergoes mean and max pooling, followed by concatenation and compression to produce a 256-dimensional triple representation: Here, X t represents the triple input which is comprised of the head, tail, and relation embeddings, and PE denotes positional encoding. The output of the triple encoding layer is input to a contrastive masking layer that identifies and preserves the most relevant features. The inner working of this layer is described in Section 3.3 . 3.1.2 Path Masking Branch In parallel with triple processing, we samples paths between the head and tail entities (path sampling is detailed in Section 4.3). Each sampled path is processed by a Path Encoding layer, which embeds each triple in the path using a position-aware neural network. The encoder applies selfattention across all triplets in the path and compresses the attended representation to 256 dimensions. The position-aware encoding ensures the path embedding maintains the sequential nature of the path: Here, X p represent the sequence of triplets in the path, and PE denotes positional encoding. The output of the path encoding layer is input to a contrastive masking layer that identifies and preserves the most relevant biological pathways. The inner working of this layer is described in Section 3.3 . 3.1.3 The Contrastive Masking Layer The Contrastive Masking layer is inspired by the information bottleneck principle of learning compressed, relevant representations. This layer serves as a learnable filter that extract the most relevant features from its input for prediction. For a given input representation X (either TripleEnc( X t ) or PathEnc( X p )), the masking layer produces both a filtered representation Z and a selection mask M through: where: g θ is a multi-layer neural network τ is a learnable threshold T is a temperature parameter ⊙ represents element-wise multiplication The selection mask M controls which features are retained, while the residual term 0.1X ensures gradient flow. g θ is a 3-layer neural network with layer normalization and GELU activation ([ 31 ]). The training objective is to learn a filtered representation Z that preserves only the information relevant for prediction from the original representation X while discarding noise. We accomplish this through contrastive learning using the InfoNCE estimator ([ 32 ]). The key idea is to minimize the mutual information between input X and filtered representation Z (compression), while maximizing the mutual information between Z and target label Y. To quantify these information relationships, the InfoNCE estimator estimates mutual information between two representations, X and Z, through contrastive learning: where f θ is a neural critic function that scores the similarity between the input X and its filtered representation Z, and N is the batch size. The masking layer processes the encoded representations to produce: These filtered representations and their corresponding masks enable both prediction and interpretation of the predictor’s decisions. The masks M t and M p highlight which features and paths were most relevant for the prediction, while Z t and Z p contain the information relevant for prediction, which is used for optimizing the training objective. The contrastive learning objectives are computed as: Where Y are the target labels. The negative term maximizes relevant selected information, while the positive term minimizes redundant information. These losses guide the Contrastive Masking layer to learn optimal feature masks, balancing between retaining relevant information for prediction while discarding noise. 3.1.4 Interaction Prediction and Loss Computation The filtered triple representations from the feature masking layer ( Z t ) are passed through a binary classification neural network. The prediction is computed as: where f classify is the classification network (a 3-layer GELU activated neural network). The Binary Cross Entropy loss between the true labels Y and predicted outputs Y ^ is calculated as: The total loss combines the prediction loss with the contrastive learning objectives: 3.2 Dataset Drug repurposing—the process of identifying new therapeutic uses for existing drugs—requires comprehensive understanding of biological interactions across multiple scales, from molecular mechanisms to clinical outcomes. While several biomedical knowledge graphs exist, including SPOKE ([ 33 ]), GARD ([ 34 ]), BioKG ([ 35 ]), and PrimeKG ([ 10 ]), we evaluate COMIC on PrimeKG due to its unique characteristics that facilitate in depth understanding of drug-disease relationships. With 22,236 disease terms consolidated into 17,080 clinically meaningful conditions, PrimeKG offers unprecedented disease coverage—1-2 orders of magnitude greater than previous datasets ([ 10 ]). This extensive disease representation is crucial for identifying novel drug applications across a wide spectrum of conditions. Furthermore, PrimeKG features 30 distinct relation types specifically designed for drug-related information, including critical relationships for repurposing such as indication, contraindication, and off-label use ([ 10 ]). These relations provide complementary signals: indication relationships serve as positive examples of validated therapeutic applications, contraindications highlight important safety boundaries and potential adverse effects, and off-label use cases offer valuable hints about promising therapeutic directions that warrant further investigation. This rich characterization of drug-disease interactions enables COMIC to learn from both successful and unsuccessful therapeutic applications while identifying promising candidates that mirror existing off-label use patterns. The remaining specialized relations enable learning indirect drug-disease interactions, which allows for more contextualized predictions and explanations. The integration of data from 20 authoritative sources, including DrugBank ([ 36 ]), DisGeNET ([ 37 ]), CTD ([ 38 ]), Gene Ontology ([ 39 ]), Human Phenotype Ontology ([ 40 ]), and Reactome ([ 41 ]), provides PrimeKG with the multi-scale biological context necessary for explainable drug repurposing ([ 10 ]). This comprehensive integration manifests in a graph structure comprising 129,375 nodes across 10 biologically relevant types, connected by 4,050,249 unique relationships, capturing the complex web of biological interactions relevant to drug repurposing, from molecular pathways to clinical manifestations. Additionally, with 90.8% of Orphanet’s 9,348 rare diseases represented, PrimeKG enables drug repurposing for rare diseases—a particularly valuable application given the limited treatment options for these conditions ([ 10 ]). These characteristics make PrimeKG an ideal foundation for developing explainable AI systems for drug repurposing, capable of both identifying novel therapeutic applications and providing mechanistic explanations for their predictions. 3.3 Evaluation Metrics To comprehensively evaluate COMIC, we employ two distinct sets of evaluation strategies: one for assessing prediction performance and another for interpretability assessment. For prediction performance, we follow standard practices from the biomedical literature [ 7 ][ 19 ][ 28 ] [ 42 ] [ 43 ] [ 44 ], using Area Under the Precision-Recall Curve (AUPRC) as the primary metric, complemented by accuracy for detailed performance analysis. The prediction evaluation metrics are summarized in Equation 1 5: where TP, TN, FP, and FN represent true positives, true negatives, false positives, and false negatives respectively. TPR denotes the true positive rate, FPR the false positive rate, and P(r) the precision at recall value r. These metrics collectively provide a robust assessment of our predictor’s ability to correctly predict drug-disease associations. Following the evaluation criteria of explainable AI (XAI) literature [ 45 ], we assess graph-based explanations using insertion and deletion faithfulness tests, which evaluate the importance of the graph’s edges in the prediction process. Additionally, we assess biological plausibility by comparing the highlighted pathways with established mechanisms of action from medical literature. We evaluate graph interpretability using the following two metrics: a. Insertion (faithfulness) : we start from an EMPTY graph and gradually INSERT top K% edges to the graph where the importance is measured by COMIC’s interpretable connections. We measure the updated AUPRC on the test set using different values of K. A better XAI method has a higher AUPRC at each threshold since it suggests that the newly added edges are informative to the graph predictions (also referred to as fidelity+). b. Deletion (faithfulness) : we start from the FULL graph and gradually DELETE top K% edges to the graph where the importance is measured by COMIC’s interpretable connections. We measure the updated AUPRC on the test set using different values of K. A better XAI method has a lower AUPRC at each threshold since it suggests that the deleted edges are necessary for the graph predictions (referred to as fidelity-). To evaluate the biological plausibility of the pathways highlighted by COMIC, we turn to medical literature and seek pathways that incorporate the indications’ mechanism of action (MOA). Specifically, we conduct a literature review to identify established biological pathways associated with each drug’s therapeutic indications and compare these with our model’s highlighted pathways. 4 Experimental Setup This section covers the proposed predictor’s experimental and implementation details. COMIC is implemented in PyTorch 2.0 for GPU-accelerated computation, with Numpy for handling vectorized operations during preprocessing and inference phases ([ 46 ]) ([ 47 ]). The web application is implemented in Django with a PostgreSQL backend. 4.1 Data Splits We utilize the data split methodology established by Huang et al., which splits PrimeKG into nine disease areas [ 7 ]. These nine splits were constructed to mirror realworld drug repurposing scenarios where new diseases have limited associated data [ 7 ]. Each holdout set is carefully curated following three strict criteria: complete exclusion of approved drugs from the training data, minimal disease similarity with the training set, and removal of biological neighbor information to limit molecular data availability [ 7 ]. Given all diseases in a given disease area, all indications and contraindications were removed from the dataset used to train machine learning models. Additionally, a fraction (5%) of the connections between biomedical entities to these diseases were removed from PrimeKG [ 7 ]. The splits encompass major therapeutic domains including adrenal gland disorders, autoimmune diseases, neurodegenerative conditions, metabolic disorders, and cardiovascular diseases, with additional splits covering anemia, diabetes, cancer, and mental health. This systematic partitioning from Huang et al. [ 7 ] ensures COMIC is tested under realistic drug repurposing conditions, where the predictor must generalize to diseases with limited prior knowledge. 4.2 Negative Sampling Strategies Negative sampling is a pivotal technique in Graph based Learning, essential for training predictors to effectively distinguish between valid and invalid relationships. The quality of negative samples significantly influences the performance of predictors, as they provide the necessary contrast to positive samples during training. However, generating high-quality negative samples poses challenges, particularly in ensuring that these samples are informative and contribute to robust learning. [ 48 ]. Various negative sampling strategies have been proposed in the literature to address these challenges. Random negative sampling is the most straightforward approach, where entities in triples are randomly replaced to create negative samples. While efficient, this method can produce trivial negatives that may not effectively challenge the predictor [ 48 ]. Model-guided and knowledge-constrained strategies leverage pre-trained models or external knowledge to generate semantically meaningful negatives, albeit with increased computational complexity [ 48 ]. Moreover, probabilistic negative sampling adjusts the sampling probability based on entity frequency, focusing on harder negatives. In our work, we implement two distinct negative sampling strategies: 1. Degree-Based Hard Negative Sampling : This approach generates negative samples by replacing either the head or tail entity in a triple, with the replacement probability proportional to the node degrees. Higher-degree nodes have a higher likelihood of being selected as replacements, resulting in more challenging negative samples that can enhance training by focusing on harder-to-predict cases. 2. Random Sampling with Fixed Destination Nodes : In this strategy, negative samples are created by maintaining the destination nodes while randomly selecting source nodes from the set of valid entities. To ensure diversity and robustness in learning, we adopt an epoch-wise regeneration strategy, where new random negative samples with fixed destinations are generated at the beginning of each training epoch. 4.3 Path Sampling Approach In large-scale knowledge graphs, comprehensive path finding between entities is often impractical due to computational constraints [ 49 ]. To address this, various path sampling strategies have been developed. One common approach is the use of meta-paths, which are predefined sequences of relation types that capture specific semantic relationships within heterogeneous graphs. This method leverages domain knowledge to guide the sampling process, focusing on meaningful and interpretable connections [ 50 ]. Alternatively, random walk-based sampling offers a more flexible approach by exploring the graph without any constraints. This allows for the discovery of novel and potentially unexpected connections, as the predictor learns path patterns directly from the data. In our study, we implement a random walk sampling strategy to explore the knowledge graph: 1. Random Walk Sampling : For each drug-disease pair, we perform random walks to sample up to 5 distinct paths, each limited to a maximum length of 5 relations. Then, 5 false paths are created by randomly corrupting the last two nodes from each path. 2. Diversity Threshold : To ensure the quality of extracted evidence, we implement a diversity threshold that prevents the collection of redundant paths. Specifically, a new path is accepted only if its overlap with previously sampled paths falls below 50%, compelling the sampler to identify varying patterns between each drug-disease pair. 4.4 Training Procedure We train the COMIC predictor using the Adam optimizer [ 51 ] with a learning rate of 0.0005. The predictor is trained for a maximum of 30 epochs with a batch size of 1024. For the contrastive masking layer, we set the temperature parameter to start at 1.0 and anneal it linearly throughout training to a minimum of 0.1, while β is fixed at 0.2. We employ early stopping with a patience of 2 epochs on the validation loss to prevent overfitting and select the best performing weights. All hyperparameters are tuned with the validation set, and we use gradient clipping with a maximum norm of 1.0 to ensure stable training. 5 Results This section illustrates the COMIC predictor’s performance under four distinct experimental settings. (1) A comprehensive comparison of COMIC against eight baseline methods and TxGNN using the established disease area splits. (2) Validation of COMIC’s real-world applicability by testing its predictions on recent FDA approvals that occurred after PrimeKG’s construction. (3) Assessment of COMIC’s interpretability through insertion and deletion experiments that demonstrate the predictor’s ability to identify biologically meaningful relationships. (4) Examination of the biological credibility of COMIC’s explanatory pathways through case studies of recently approved drug-disease pairs. 5.1 Predictive Performance COMIC’s predictive performance is compared against eight baseline methods and TxGNN using the disease area splits and results reported by [ 7 ] (described in Section 4.2.1). The baseline methods span network statistical techniques (Kullback–Leibler divergence, Jensen–Shannon divergence), graph-theoretical approaches (network proximity, diffusion state distance), modern GNN architectures (RGCN, HGT, HAN), and language models (BioBERT) [ 52 ] [ 53 ] [ 54 ] [ 17 ] [ 55 ] [ 56 ] [ 57 ] [ 58 ]. To ensure robust comparison, we first conducted an analysis of negative sampling strategies. While Huang et al.[ 7 ] employed both random and degree-based sampling in their work, they did not specify which approach was used for their final reported results. Given this ambiguity, we evaluated both approaches through validation on recent FDA-approved drugs. We found that degree-based hard negative sampling outperformed random generation by an average margin of 10.2% in ranking correct indications higher. This superior performance better reflects real-world drug repurposing scenarios by forcing the predictor to correctly distinguish between therapeutically similar but distinct drug-disease pairs. More importantly, by choosing degree-based hard negative sampling for our comparative analysis, we ensure our reported performance metrics represent conservative estimates relative to potentially simpler random sampling approaches. All our results reported in this section use degree-based hard negative sampling and are averaged across five random seeds (1-5), providing a robust and conservative comparison baseline. COMIC demonstrated consistently high performance across all nine disease areas for both indication and contraindication prediction tasks ( Table 1 and 2). For indications, COMIC achieved AUPRC scores ranging from 0.88 to 0.94, surpassing the state-of-the-art ( Table 1 ). Notably, while the state-of-the-art [ 7 ] showed strong performance in specific areas (AUPRC 0.95 for adrenal gland, anemia, and neurodegenerative diseases), COMIC maintained more consistent performance across all disease categories, with an average improvement of 4.29% over TxGNN. This consistency is particularly evident in challenging areas like cardiovascular diseases, where COMIC achieved an AUPRC of 0.88 compared to TxGNN’s 0.64, representing a 37.8% improvement. For contraindications, COMIC outperformed existing methods across all disease areas, with AUPRC scores ranging from 0.88 to 0.96 ( Table 2 ). The predictor achieved particularly strong results in diabetes (AUPRC: 0.95) and metabolic disorders (AUPRC: 0.95), areas where accurate contraindication prediction is crucial for patient safety. The performance gap was again most pronounced in cardiovascular diseases, where COMIC maintained robust performance (AUPRC: 0.88) while other methods showed significant degradation. Again, COMIC maintained more consistent performance across all disease categories, with an average improvement of 14.82% over TxGNN. COMIC overall achieved an average gain of 9.55% across both interaction types. These results establish COMIC’s predictive power in a test setting. View this table: View inline View popup Download powerpoint Table 1 Performance in predicting drug indications. Reported is AUPRC. View this table: View inline View popup Download powerpoint Table 2 Performance in predicting drug contraindications. Reported is AUPRC. 5.2 Evaluation of COMIC on recent FDA-approvals The next step is to validate COMIC’s real-world applicability. For this, we test the predictor on 30 repurposed drug-disease pairs that received FDA approval between 2022-2025. These approvals occurred after the knowledge graph’s construction, ensuring that no relationships between these drug-disease pairs existed in the training data. Highlighted in Table 3 , this evaluation provides a stringent test of the predictor’s ability to identify drug repurposing candidates. View this table: View inline View popup Download powerpoint Table 3 COMIC’s performance on recent FDA-Approved Repurposed Indications. The percentile indicates how high the drug was ranked relative to all possible indications for the disease. Despite achieving exceptional AUPRC scores on the test set, COMIC’s performance on completely novel FDA approvals revealed important real-world challenges. The predictor successfully predicted 21 out of 30 (70%) approved indications, with 10 of these predictions ranking above the 90th percentile ( Table 3 ). This notable performance gap between testing and prospective evaluation underscores the inherent difficulty of predicting therapeutic applications, particularly when mechanistic understanding is still evolving for most diseases. COMIC still demonstrated strong predictive ability for several therapeutic areas. For autoimmune conditions, Ustekinumab for Psoriatic Arthritis ranked in the 97th percentile, while for metabolic disorders, Tirzepatide for Type 2 Diabetes ranked in the 95th percentile. Aflibercept showed consistently high rankings across multiple indications: diabetic macular edema (96th percentile), macular degeneration (94th percentile), and diabetic retinopathy (92nd percentile). To further investigate the gap in performance, we compared COMIC’s performance against TxGNN using the TxGNNExplorer ( txgnn.org ) [ 7 ], which provided TxGNN scores for 18 out of the 30 drug-disease pairs as shown in Table 3 . The comparison reveals COMIC’s superior discrimination ability (scoring higher for 12 out of 18 pairs), suggesting our predictor captures complex molecular interactions better in the context of these pairs. Where COMIC struggled most was with certain cancers and novel therapeutic mechanisms. For instance, Nivolumab received varied predictions across different cancer types, ranging from high confidence for Colorectal Cancer (86th percentile) to low confidence for Urothelial Carcinoma (1.8th percentile) and Renal Cell Carcinoma (8.5th percentile). Similarly, recent Alzheimer’s disease therapy Lecanemab received a low ranking (1.8th percentile), highlighting challenges in neurodegenerative diseases where their mechanisms are still being understood [ 59 ]. These results reveal both COMIC’s strengths and limitations in real-world drug repurposing scenarios. While the 70% prediction rate demonstrates substantial practical utility, the performance differential between the test set and recent approvals highlights opportunities for further refinement. 5.3 Evaluation of COMIC’s Interpretability To assess COMIC’s interpretability, we conduct insertion and deletion evaluation described in Section 3.3 . In the insertion experiment, as shown in Figure 3 , the AUPRC steadily increased—from 0.67 at 10% insertion up to 0.99 when all edges were inserted—demonstrating that the edges identified as influential indeed contribute substantially to COMIC’s predictive performance. Conversely, the deletion test started with the full graph and gradually removed edges ranked by their importance. As shown in Figure 2 , the resulting deletion curve reveals a marked decrease in AUPRC at each step (from approximately 0.99 with no removal to 0.62 at 90% deletion), highlighting that even modest removals of highly ranked edges significantly impair COMIC’s outputs. Download figure Open in new tab Fig. 2 Insertion Performance Evaluation Download figure Open in new tab Fig. 3 Deletion Performance Evaluation The initial steep decrease in deletion and the initial steep increase in insertion conform to the ideal in these evaluation metrics, confirming that COMIC’s interpretable connections are faithful to the predictor’s underlying decision-making process. Together, these trends indicate that the most highly-ranked edges are truly critical for accurate predictions as indicated by the rapid initial degradation during deletion and high initial gains during insertion. This demonstrates that COMIC’s interpretability is transparent and robust in identifying the edges most crucial to the predictor’s decision process. 5.4 Mechanistic Pathways Validation We now examine the biological credibility of its explanatory pathways. We consider the indication pairs in Table 3 for stringent testing since they were approved after PrimeKG’s construction. In the case of Ocrelizumab for Multiple Sclerosis, COMIC identified several pathways that directly link its target, MS4A1 (CD20), to key immune mediators ([ 60 ]). Notably, several valid pathways traverse through molecules such as HLA-DRA, CD40, NF-KB, and IL7R—elements that are well documented in the pathogenesis of multiple sclerosis and align closely with the drug’s known MOA of B cell depletion and subsequent modulation of T cell responses ([ 61 ]) ([ 62 ]) ([ 63 ]). These mechanistic explanations are further supported by literature that underscores the importance of MHC class II interactions and costimulatory signals in mediating autoimmunity ([ 64 ]). Similarly, for Atezolizumab in Melanoma, COMIC revealed pathways that directly connect CD274 (PD-L1) to known melanoma drivers ([ 65 ]). In particular, pathways such as the MTOR–AKT1 cascade reflect established oncogenic processes in melanoma ([ 66 ]). Additionally, a link through SLC39A11 to PMEL (gp100) directly ties to a melanocyte differentiation antigen and a key target in melanoma immunotherapy ([ 67 ]). These paths provide mechanistically plausible explanations that resonate with current understanding of melanoma biology and the rationale behind immune checkpoint inhibition. In contrast, for Bortezomib and Mantle Cell Lymphoma, the predictor predominantly returned paths based on drug effects—for instance, adverse effects such as dyssynergia, hyperuricemia, connect to drugs, which then eventually tie to other phenotypes like fatigue and, indirectly, to mantle cell lymphoma. Such paths do not capture the primary mechanism of action of Bortezomib, but instead appear to be driven by indirect associations present in the dataset. In cases like this where the underlying molecular mechanisms are not well represented in the knowledge graph, COMIC did not find any meaningful mechanistic pathways. For instance, for Alzheimer’s disease and lecanemab, for asthma and tezepelumab, or for hemophilia and concizumab, the predictor’s explanations were either absent or based on indirect disease associations rather than direct MOA-related pathways. These examples underscore that while COMIC is capable of retrieving highly plausible mechanistic paths when the underlying biology is present in the dataset, its explanations for indications lacking comprehensive mechanistic data tend to rely on indirect associations. This reliance on indirect associations when insufficient evidence is not in the dataset for certain indications emphasizes the need for further enrichment of the underlying knowledge graph and continued refinement of our interpretability methods. 5.5 The COMIC Web Application After extensive evaluation of COMIC’s capabilities, we have deployed the predictor as a web application ( https://drp.opendfki.de/ ) that provides predictions and their explanations for drug indications and contraindications. Users can select/search diseases, configure pathway parameters, and search for specific drugs. Results are presented in a sortable table with confidence scores for each drug-disease pair. The application also visualizes the pathway explanations connecting drugs to diseases through interactive network diagrams, helping researchers understand the potential mechanisms of action. This web applications aims to accelerate drug repurposing research by making COMIC’s predictions accessible to the broader scientific community, serving as a valuable resource for drug development researchers, clinicians exploring alternative therapies, and computational biologists investigating disease mechanisms. 6 Conclusion The core contribution of this work lies in the development of a novel predictor, COMIC, that enables the systematic identification of potential new indications for existing drugs, with a particular focus on diseases that are characterized by a lack of effective therapeutic options. This approach offers a significant advancement in addressing the unmet medical needs associated with these conditions. COMIC achieves a 4.29% average improvement in indication prediction over the state-of-the-art in a test setting. The superior performance of the COMIC predictor, compared to existing approaches, is driven by its innovative feature and path masking branches. The feature masking branch successfully identifies critical drug-disease interaction signatures, while the path masking branch is capable of highlighting mechanistic pathways, together providing clinicians with transparent reasoning for COMIC’s predictions. COMIC’s real-world applicability is validated through its correct identification of 70% of recently approved therapeutic indications, demonstrating considerable practical utility while maintaining end-to-end interpretability. But the evaluation on recent FDA approvals also revealed important limitations, particularly for diseases with evolving molecular understanding such as certain cancers and neurodegenerative conditions. These challenges highlight the critical need for richer knowledge graphs with more comprehensive mechanistic data, especially for rare diseases where treatment options remain limited. Future research should focus on expanding the biological pathway representations within these graphs and refining interpretable mechanisms to better capture complex therapeutic relationships. By addressing these data gaps and continuing to enhance interpretability, we can further bridge the divide between computational predictions and clinical adoption of AI-driven drug repurposing. Supplementary information Accuracy results on train, valid, and test sets of each of the nine disease splits have been attached as supplementary material. Declarations Funding: There is no funding source. Conflict of interest/Competing interests: There is no author’s conflict of interest. Ethics approval and consent to participate: Research is conducted responsibly, with respect for participants’ autonomy, dignity, and well-being Consent for publication: Authors provide consent for their work to be shared in journal. Data availability: The dataset supporting the conclusions of this article is available in the Harvard Dataverse repository. And the data splits can be accessed from the TxGNN GitHub Repository. Materials availability: All material is available in the paper. Code availability: The predictor is publicly available at the following URL: https://drp.opendfki.de/ Authors contribution : – Naafey Aamer: Conceptualization, Methodology, Validation, Investigation, Formal analysis, Visualization, Data curation, Manuscript writing – Muhammad Nabeel Asim: Conceptualization, Methodology, Validation, Manuscript writing, review and editing manuscript, Supervision. – Andreas Dengel: Supervision, Reviewed the article and performed final editing Footnotes https://drp.opendfki.de/ References [1]. ↵ U.S. Food and Drug Administration: Drugs@FDA Glossary of Terms . Accessed: 2025-03-26 ( 2024 ). https://www.fda.gov/drugs/drug-approvals-and-databases/ drugsfda-glossary-terms [2]. ↵ Abdelsayed , M. , Kort , E.J. , Jovinge , S. , Mercola , M .: Repurposing drugs to treat cardiovascular disease in the era of precision medicine . Nature Reviews Cardiology 19 , 751 – 764 ( 2022 ) OpenUrl PubMed [3]. ↵ Sahragardjoonegani , B. , Beall , R.F. , Kesselheim , A.S. , Hollis , A .: Repurposing existing drugs for new uses: a cohort study of the frequency of fda-granted new indication exclusivities since 1997 . Journal of Pharmaceutical Policy and Practice 14 , 3 ( 2021 ) [4]. ↵ Sardana , D. , et al : Drug repositioning for orphan diseases . Briefings in Bioinformatics 12 , 346 – 356 ( 2011 ) OpenUrl CrossRef PubMed [5]. ↵ Attia , G.M. , Almouteri , M.M. , Alnakhli , F.T .: Role of metformin in polycystic ovary syndrome (pcos)-related infertility . Cureus 15 ( 8 ), 44493 ( 2023 ) doi: 10.7759/cureus.44493 OpenUrl CrossRef [6]. ↵ Richardson , P. , Hideshima , T. , Anderson , K .: Thalidomide in multiple myeloma . Biomedicine & Pharmacotherapy 56 ( 3 ), 115 – 128 ( 2002 ) doi: 10.1016/S0753-3322(02)00168-3 OpenUrl CrossRef PubMed [7]. ↵ Huang , K. , Chandak , P. , Wang , Q. , et al : A foundation model for cliniciancentered drug repurposing . Nature Medicine 30 , 3601 – 3613 ( 2024 ) doi: 10.1038/s41591-024-03233-x OpenUrl CrossRef [8]. ↵ Saber-Ayad , M. , Hammoudeh , S. , Abu-Gharbieh , E. , Hamoudi , R. , Tarazi , H. , Al-Tel , T.H. , Hamid , Q .: Current status of baricitinib as a repurposed therapy for covid-19 . Pharmaceuticals 14 ( 7 ), 680 ( 2021 ) doi: 10.3390/ph14070680 OpenUrl CrossRef PubMed [9]. ↵ Paiva , H.S. , Filho , I.J.Z. , Cais , C.F.D.S. : Using prazosin to treat posttraumatic stress disorder and associations: A systematic review . Psychiatry Investigation 18 ( 5 ), 365 – 372 ( 2021 ) doi: 10.30773/pi.2020.0411 OpenUrl CrossRef PubMed [10]. ↵ Chandak , P. , Huang , K. , Zitnik , M .: Building a knowledge graph to enable precision medicine . Scientific Data 10 , 67 ( 2023 ) [11]. ↵ Li , M.M. , Huang , K. , Zitnik , M .: Graph representation learning in biomedicine and healthcare . Nature Biomedical Engineering 6 , 1353 – 1369 ( 2022 ) OpenUrl PubMed [12]. ↵ Aamer , N. , Asim , M.N. , Bhatti , A.I. , Dengel , A. : Bind: Large-scale biological interaction network discovery through knowledge graph-driven machine learning . bioRxiv ( 2025 ) doi: 10.1101/2025.01.15.633109 https://www.biorxiv.org/content/early/2025/01/19/2025.01.15.633109.full.pdf OpenUrl Abstract / FREE Full Text [13]. ↵ Tiddi , I. , Schlobach , S. : Knowledge graphs as tools for explainable machine learning: A survey . Artificial Intelligence 302 , 103627 ( 2022 ) doi: 10.1016/ j.artint.2021.103627 OpenUrl CrossRef PubMed [14]. ↵ Vora , L.K. , Gholap , A.D. , Jetha , K. , Thakur , R.R.S. , Solanki , H.K. , Chavda , V.P .: Artificial intelligence in pharmaceutical technology and drug delivery design . Pharmaceutics 15 ( 7 ), 1916 ( 2023 ) doi: 10.3390/pharmaceutics15071916 OpenUrl CrossRef PubMed [15]. ↵ Chan , S.Y. , Loscalzo , J .: The emerging paradigm of network medicine in the study of human disease . Circulation Research 111 ( 3 ), 359 – 74 ( 2012 ) doi: 10.1161/CIRCRESAHA.111.258541 OpenUrl Abstract / FREE Full Text [16]. ↵ Cha , Y. , Erez , T. , Reynolds , I.J. , Kumar , D. , Ross , J. , Koytiger , G. , Kusko , R. , Zeskind , B. , Risso , S. , Kagan , E. , Papapetropoulos , S. , Grossman , I. , Laifenfeld , D .: Drug repurposing from the perspective of pharmaceutical companies . British Journal of Pharmacology 175 ( 2 ), 168 – 180 ( 2018 ) doi: 10.1111/bph.13798 OpenUrl CrossRef PubMed [17]. ↵ Zitnik , M. , Agrawal , M. , Leskovec , J .: Modeling polypharmacy side effects with graph convolutional networks . Bioinformatics 34 , 457 – 466 ( 2018 ) OpenUrl [18]. ↵ Saraei , P. , Asadi , I. , Kakar , M.A. , Moradi-Kor , N .: The beneficial effects of metformin on cancer prevention and therapy: a comprehensive review of recent advances . Cancer Management and Research 11 , 3295 – 3313 ( 2019 ) doi: 10.2147/CMAR.S200059 OpenUrl CrossRef PubMed [19]. ↵ Ma , C. , Zhou , Z. , Liu , H. , Koslicki , D. : Kgml-xdtd: a knowledge graph–based machine learning framework for drug treatment prediction and mechanism description . GigaScience 12 , 057 ( 2023 ) doi: 10.1093/gigascience/giad057 OpenUrl CrossRef [20]. ↵ Liu , Y. , Hildebrandt , M. , Joblin , M. , Ringsquandl , M. , Raissouni , R. , Tresp , V. : Neural Multi-Hop Reasoning With Logical Rules on Biomedical Knowledge Graphs ( 2021 ). https://arxiv.org/abs/2103.10367 [21]. ↵ Das , R. , Dhuliawala , S. , Zaheer , M. , Vilnis , L. , Durugkar , I. , Krishnamurthy , A. , Smola , A. , McCallum , A. : Go for a walk and arrive at the answer: Reasoning over paths in knowledge bases using reinforcement learning . In: ICLR ( 2018 ) [22]. ↵ Ott , S. , Meilicke , C. , Samwald , M. : SAFRAN: An interpretable, rulebased link prediction method outperforming embedding models . In: 3rd Conference on Automated Knowledge Base Construction ( 2021 ). https://openreview.net/forum?id=jCt9S 3w S9 [23]. ↵ Lee , C.Y. , Chen , Y.-P.P .: New insights into drug repurposing for covid-19 using deep learning . IEEE Transactions on Neural Networks and Learning Systems 32 ( 11 ), 4770 – 4780 ( 2021 ) doi: 10.1109/TNNLS.2021.3111745 OpenUrl CrossRef [24]. ↵ Fermaglich , L.J. , Miller , K.L .: A comprehensive study of the rare diseases and conditions targeted by orphan drug designations and approvals over the forty years of the orphan drug act . Orphanet Journal of Rare Diseases 18 , 1 – 8 ( 2023 ) OpenUrl PubMed [25]. ↵ Jiménez , A. , Merino , M.J. , Parras , J. , Zazo , S. : Explainable drug repurposing via path based knowledge graph completion . Scientific Reports 14 ( 1 ), 16587 ( 2024 ) doi: 10.1038/s41598-024-67163-x OpenUrl CrossRef PubMed [26]. ↵ Qu , M. , Chen , J. , Xhonneux , L.-P. , Bengio , Y. , Tang , J. : Rnnlogic: Learning logic rules for reasoning on knowledge graphs . arXiv abs/2010.04029 ( 2020 ) [27]. ↵ Ying , R. , Bourgeois , D. , You , J. , Zitnik , M. , Leskovec , J. : GNNExplainer: Generating Explanations for Graph Neural Networks ( 2019 ). https://arxiv.org/abs/1903.03894 [28]. ↵ Perdomo-Quinteiro , P. , Wolstencroft , K. , Roos , M. , Queralt-Rosinach , N .: Knowledge graphs and explainable ai for drug repurposing on rare diseases ( 2024 ) doi: 10.1101/2024.10.17.618804 OpenUrl Abstract / FREE Full Text [29]. ↵ Schlichtkrull , M.S. , Cao , N.D. , Titov , I. : Interpreting graph neural networks for {nlp} with differentiable edge masking . In: International Conference on Learning Representations ( 2021 ). https://openreview.net/forum?id=WznmQa42ZAx [30]. ↵ Teh , Y.W. Glorot , X. , Bengio , Y. : Understanding the difficulty of training deep feedforward neural networks . In: Teh , Y.W. , Titterington , M. (eds.) Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics. Proceedings of Machine Learning Research , vol. 9, pp. 249 – 256 . PMLR, Chia Laguna Resort , Sardinia, Italy ( 2010 ). https://proceedings.mlr.press/v9/glorot10a.html [31]. ↵ Hendrycks , D. , Gimpel , K. : Gaussian Error Linear Units (GELUs) ( 2023 ). https://arxiv.org/abs/1606.08415 [32]. ↵ Oord , A. , Li , Y. , Vinyals , O. : Representation Learning with Contrastive Predictive Coding ( 2019 ). https://arxiv.org/abs/1807.03748 [33]. ↵ Morris , J.H. , Soman , K. , Akbas , R.E. , Zhou , X. , Smith , B. , Meng , E.C. , Huang , C.C. , Cerono , G. , Schenk , G. , Rizk-Jackson , A. , Harroud , A. , Sanders , L. , Costes , S.V. , Bharat , K. , Chakraborty , A. , Pico , A.R. , Mardirossian , T. , Keiser , M. , Tang , A. , Hardi , J. , Shi , Y. , Musen , M. , Israni , S. , Huang , S. , Rose , P.W. , Nelson , C.A. , Baranzini , S.E .: The scalable precision medicine open knowledge engine (SPOKE): a massive knowledge graph of biomedical information . Bioinformatics 39 ( 2 ), 080 ( 2023 ) doi: 10.1093/bioinformatics/btad080 OpenUrl CrossRef [34]. ↵ Zhu , Q. , Nguyen , D.-T. , Grishagin , I. , et al : An integrative knowledge graph for rare diseases, derived from the Genetic and Rare Diseases Information Center (GARD) . Journal of Biomedical Semantics 11 , 13 ( 2020 ) doi: 10.1186/s13326-020-00232-y OpenUrl CrossRef PubMed [35]. ↵ Walsh , B. , Mohamed , S.K. , Nováček , V. : Biokg: A knowledge graph for relational learning on biological data . In: Proceedings of the 29th ACM International Conference on Information & Knowledge Management. CIKM’20 , pp. 3173 – 3180 . Association for Computing Machinery, New York , NY, USA ( 2020 ). doi: 10.1145/3340531.3412776 . 10.1145/3340531.3412776 OpenUrl CrossRef [36]. ↵ Knox , C. , Wilson , M. , Klinger , C.M. , Franklin , M. , Oler , E. , Wilson , A. , Pon , A. , Cox , J. , Chin , N.E.L. , Strawbridge , S.A. , Garcia-Patino , M. , Kruger , R. , Sivakumaran , A. , Sanford , S. , Doshi , R. , Khetarpal , N. , Fatokun , O. , Doucet , D. , Zubkowski , A. , Yahya Rayat , D. , Jackson , H. , Harford , K. , Anjum , A. , Zakir , M. , Wang , F. , Tian , S. , Lee , B. , Liigand , J. , Peters , H. , Wang , R.Q.R. , Nguyen , T. , So , D. , Sharp , M. , Silva , R. , Gabriel , C. , Scantlebury , J. , Jasinski , M. , Ackerman , D. , Jewison , T. , Sajed , T. , Gautam , V. , Wishart , D.S .: DrugBank 6.0: the Drug-Bank Knowledgebase for 2024 . Nucleic Acids Research 52 ( D1 ), 1265 – 1275 ( 2024 ) doi: 10.1093/nar/gkad976 OpenUrl CrossRef [37]. ↵ Piñero , J. , Ramírez-Anguita , J.M. , Saüch-Pitarch , J. , Ronzano , F. , Centeno , E. , Sanz , F. , Furlong , L.I. : The DisGeNET knowledge platform for disease genomics: 2019 update . Nucleic Acids Research 48 ( D1 ), 845 – 855 ( 2020 ) doi: 10.1093/nar/gkz1021 OpenUrl CrossRef PubMed [38]. ↵ Davis , A.P. , Grondin , C.J. , Johnson , R.J. , Sciaky , D. , Wiegers , J. , Wiegers , T.C. , Mattingly , C.J .: Comparative Toxicogenomics Database (CTD): update 2021 . Nucleic Acids Research 49 ( D1 ), 1138 – 1143 ( 2021 ) doi: 10.1093/nar/gkaa891 OpenUrl CrossRef [39]. ↵ The Gene Ontology Consortium , Aleksander , S.A. , Balhoff , J. , Carbon , S. , Cherry , J.M. , Drabkin , H.J. , Ebert , D. , Feuermann , M. , Gaudet , P. , Harris , N.L. , Hill , D.P. , Lee , R. , Mi , H. , Moxon , S. , Mungall , C.J. , Muruganugan , A. , Mushayahama , T. , Sternberg , P.W. , Thomas , P.D. , Van Auken , K. , Ramsey , J. , Siegele , D.A. , Chisholm , R.L. , Fey , P. , Aspromonte , M.C. , Nugnes , M.V. , Quaglia , F. , Tosatto , S. , Giglio , M. , Nadendla , S. , Antonazzo , G. , Attrill , H. , Santos , G. , Marygold , S. , Strelets , V. , Tabone , C.J. , Thurmond , J. , Zhou , P. , Ahmed , S.H. , Asanitthong , P. , Buitrago , D.L. , Erdol , M.N. , Gage , M.C. , Kadhum , M.A. , Li , K.Y.C. , Long , M. , Michalak , A. , Pesala , A. , Pritazahra , A. , Saverimuttu , S.C.C. , Su , R. , Thurlow , K.E. , Lovering , R.C. , Logie , C. , Oliferenko , S. , Blake , J. , Christie , K. , Corbani , L. , Dolan , M.E. , Drabkin , H.J. , Hill , D.P. , Ni , L. , Sitnikov , D. , Smith , C. , Cuzick , A. , Seager , J. , Cooper , L. , Elser , J. , Jaiswal , P. , Gupta , P. , Jaiswal , P. , Naithani , S. , Lera-Ramirez , M. , Rutherford , K. , Wood , V. , De Pons , J.L. , Dwinell , M.R. , Hayman , G.T. , Kaldunski , M.L. , Kwitek , A.E. , Laulederkind , S.J.F. , Tutaj , M.A. , Vedi , M. , Wang , S.-J. , D’Eustachio , P. , Aimo , L. , Axelsen , K. , Bridge , A. , Hyka-Nouspikel , N. , Morgat , A. , Aleksander , S.A. , Cherry , J.M. , Engel , S.R. , Karra , K. , Miyasato , S.R. , Nash , R.S. , Skrzypek , M.S. , Weng , S. , Wong , E.D. , Bakker , E. , Berardini , T.Z. , Reiser , L. , Auchincloss , A. , Axelsen , K. , Argoud-Puy , G. , Blatter , M.-C. , Boutet , E. , Breuza , L. , Casals-Casas , C. , Coudert , E. , Estreicher , A. , Famiglietti , M.L. , Feuermann , M. , Gos , A. , Gruaz-Gumowski , N. , Hulo , C. , Hyka-Nouspikel , N. , Jungo , F. , Le Mercier , P. , Lieberherr , D. , Masson , P. , Morgat , A. , Pedruzzi , I. , Pourcel , L. , Poux , S. , Rivoire , C. , Sundaram , S. , Bateman , A. , Bowler-Barnett , E. , Bye-A-Jee , H. , Denny , P. , Ignatchenko , A. , Ishtiaq , R. , Lock , A. , Lussi , Y. , Magrane , M. , Martin , M.J. , Orchard , S. , Raposo , P. , Speretta , E. , Tyagi , N. , Warner , K. , Zaru , R. , Diehl , A.D. , Lee , R. , Chan , J. , Diamantakis , S. , Raciti , D. , Zarowiecki , M. , Fisher , M. , James-Zorn , C. , Ponferrada , V. , Zorn , A. , Ramachandran , S. , Ruzicka , L. , Westerfield , M. : The Gene Ontology knowledgebase in 2023 . Genetics 224 ( 1 ), 031 ( 2023 ) doi: 10.1093/genetics/iyad031 OpenUrl CrossRef PubMed [40]. ↵ Gargano , M.A. , et al : The Human Phenotype Ontology in 2024: phenotypes around the world . Nucleic Acids Research 52 ( D1 ), 1333 – 1346 ( 2024 ) doi: 10.1093/nar/gkad1005 OpenUrl CrossRef [41]. ↵ Fabregat , A. , Jupe , S. , Matthews , L. , Sidiropoulos , K. , Gillespie , M. , Garapati , P. , Haw , R. , Jassal , B. , Korninger , F. , May , B. , Milacic , M. , Roca , C.D. , Rothfels , K. , Sevilla , C. , Shamovsky , V. , Shorser , S. , Varusai , T. , Viteri , G. , Weiser , J. , Wu , G. , Stein , L. , Hermjakob , H. , D’Eustachio , P .: The Reactome Pathway Knowledgebase . Nucleic Acids Research 46 ( D1 ), 649 – 655 ( 2018 ) doi: 10.1093/nar/gkx1132 OpenUrl CrossRef [42]. ↵ Asim , M.N. , Fazeel , A. , Ibrahim , M.A. , Dengel , A. , Ahmed , S. : Mp-vhppi: Meta predictor for viral host protein-protein interaction prediction in multiple hosts and viruses . Frontiers in Medicine Volume 9 - 2022 ( 2022 ) doi: 10.3389/fmed.2022.1025887 OpenUrl CrossRef [43]. ↵ Asim , M.N. , Ibrahim , M.A. , Malik , M.I. , Dengel , A. , Ahmed , S. : Lgca-vhppi: A local-global residue context aware viral-host protein-protein interaction predictor . PLOS ONE 17 ( 7 ), 0270275 ( 2022 ) doi: 10.1371/journal.pone.0270275 OpenUrl CrossRef [44]. ↵ Nabeel Asim , M. , Ali Ibrahim , M. , Fazeel , A. , Dengel , A. , Ahmed , S. : Dnamp: a generalized dna modifications predictor for multiple species based on powerful sequence encoding method . Briefings in Bioinformatics 24 ( 1 ), 546 ( 2022 ) doi: 10.1093/bib/bbac546 https://academic.oup.com/bib/article-pdf/24/1/bbac546/48782667/supplementary-file-2bbac546.pdf OpenUrl CrossRef [45]. ↵ Muresan , S. Chan , C.S. , Kong , H. , Guanqing , L. : A comparative study of faithfulness metrics for model interpretability methods . In: Muresan , S. , Nakov , P. , Villavicencio , A. (eds.) Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pp. 5029 – 5038 . Association for Computational Linguistics , Dublin, Ireland ( 2022 ). doi: 10.18653/v1/2022.acl-long.345 . https://aclanthology.org/2022.acl-long.345/ OpenUrl CrossRef [46]. ↵ Paszke , A. , Gross , S. , Massa , F. , Lerer , A. , Bradbury , J. , Chanan , G. , Killeen , T. , Lin , Z. , Gimelshein , N. , Antiga , L. , Desmaison , A. , Köpf , A. , Yang , E. , DeVito , Z. , Raison , M. , Tejani , A. , Chilamkurthy , S. , Steiner , B. , Fang , L. , Bai , J. , Chintala , S. : PyTorch: An Imperative Style, High-Performance Deep Learning Library ( 2019 ). https://arxiv.org/abs/1912.01703 [47]. ↵ Harris , C.R. , Millman , K.J. , Walt , S.J. , et al : Array programming with NumPy . Nature 585 , 357 – 362 ( 2020 ) doi: 10.1038/s41586-020-2649-2 OpenUrl CrossRef PubMed [48]. ↵ Madushanka , T. , Ichise , R. : Negative Sampling in Knowledge Graph Representation Learning: A Review ( 2024 ). https://arxiv.org/abs/2402.19195 [49]. ↵ Cai , Y. , Liu , S. , Zheng , W. , Lin , X .: Towards generating hop-constrained s-t simple path graphs . Proc. ACM Manag. Data 1 ( 1 ) ( 2023 ) doi: 10.1145/3588915 OpenUrl CrossRef [50]. ↵ Liu , A. , Chen , J. , Du , R. , Wu , C. , Feng , Y. , Li , T. , Ma , J. : HeteroSample: Metapath Guided Sampling for Heterogeneous Graph Representation Learning ( 2024 ). https://arxiv.org/abs/2411.07022 [51]. ↵ Kingma , D.P. , Ba , J. : Adam: A Method for Stochastic Optimization ( 2017 ). https://arxiv.org/abs/1412.6980 [52]. ↵ Gysi , D.M. , et al : Network medicine framework for identifying drug-repurposing opportunities for Covid-19 . Proceedings of the National Academy of Sciences of the United States of America 118 , 2025581118 ( 2021 ) OpenUrl CrossRef [53]. ↵ Guney , E. , Menche , J. , Vidal , M. , Barabásii , A.-L. : Network-based in silico drug efficacy screening . Nature Communications 7 , 1 – 13 ( 2016 ) OpenUrl CrossRef [54]. ↵ Cao , M. , et al : Going the distance for protein function prediction: a new distance metric for protein interaction networks . PLoS ONE 8 , 76339 ( 2013 ) OpenUrl [55]. ↵ Schlichtkrull , M. , et al : Modeling relational data with graph convolutional networks . In: Gangemi , A. , et al. (eds.) The Semantic Web: 15th International Conference, ESWC 2018 , pp. 593 – 607 . Springer ,( 2018 ) [56]. ↵ Hu , Z. , Dong , Y. , Wang , K. , Sun , Y. : Heterogeneous graph transformer . In: Huang , Y. , et al. (eds.) Proceedings of the World Wide Web Conference 2020 , pp. 2704 – 2710 . Association for Computing Machinery ,( 2020 ) [57]. ↵ Wang , X. , Ji , H. , Shi , C. , Wang , B. , Ye , Y. , Cui , P. , Yu , P.S. : Heterogeneous graph attention network . In: Liu , L. , White , R. , et al. (eds.) Proceedings of the World Wide Web Conference 2019 , pp. 2022 – 2032 . Association for Computing Machinery ,( 2019 ) [58]. ↵ Lee , J. , et al : BioBERT: a pre-trained biomedical language representation model for biomedical text mining . Bioinformatics 36 , 1234 – 1240 ( 2020 ) OpenUrl CrossRef PubMed [59]. ↵ Gadhave , D.G. , Sugandhi , V.V. , Jha , S.K. , Nangare , S.N. , Gupta , G. , Singh , S.K. , Dua , K. , Cho , H. , Hansbro , P.M. , Paudel , K.R. : Neurodegenerative disorders: Mechanisms of degeneration and therapeutic approaches with their clinical relevance . Ageing Research Reviews 99 , 102357 ( 2024 ) doi: 10.1016/ j.arr.2024.102357 OpenUrl CrossRef PubMed [60]. ↵ Florou , D. , Katsara , M. , Feehan , J. , Dardiotis , E. , Apostolopoulos , V. : Anti-CD20 agents for multiple sclerosis: Spotlight on ocrelizumab and ofatumumab . Brain Sciences 10 ( 10 ), 758 ( 2020 ) doi: 10.3390/brainsci10100758 OpenUrl CrossRef PubMed [61]. ↵ Morrison , B.A. , et al : Multiple sclerosis risk markers in HLA-DRA, HLA-C, and IFNG genes are associated with sex-specific childhood leukemia risk . Autoimmunity 43 ( 8 ), 690 – 7 ( 2010 ) doi: 10.3109/08916930903567492 OpenUrl CrossRef PubMed [62]. ↵ Aarts , S.A.B.M. , Seijkens , T.T.P. , Dorst , K.J.F. , Dijkstra , C.D. , Kooij , G. , Lutgens , E .: The CD40-CD40L dyad in experimental autoimmune encephalomyelitis and multiple sclerosis . Frontiers in Immunology 8 , 1791 ( 2017 ) doi: 10.3389/fimmu.2017.01791 OpenUrl CrossRef PubMed [63]. ↵ Lamb , Y.N. : Ocrelizumab: A review in multiple sclerosis . Drugs 82 ( 3 ), 323 – 334 ( 2022 ) doi: 10.1007/s40265-022-01672-9 OpenUrl CrossRef PubMed [64]. ↵ Ishina , I.A. , Zakharova , M.Y. , Kurbatskaia , I.N. , Mamedov , A.E. , Belogurov , A.A.J. , Gabibov , A.G .: MHC class II presentation in autoimmunity . Cells 12 ( 2 ), 314 ( 2023 ) doi: 10.3390/cells12020314 OpenUrl CrossRef [65]. ↵ Shah , N.J. , Kelly , W.J. , Liu , S.V. , Choquette , K. , Spira , A .: Product review on the Anti-PD-L1 antibody atezolizumab . Human Vaccines & Immunotherapeutics 14 ( 2 ), 269 – 276 ( 2018 ) doi: 10.1080/21645515.2017.1403694 OpenUrl CrossRef PubMed [66]. ↵ Peng , Y. , Wang , Y. , Zhou , C. , Mei , W. , Zeng , C. : PI3K/Akt/mTOR pathway and its role in cancer therapeutics: Are we making headway? Frontiers in Oncology 12 , 819128 ( 2022 ) doi: 10.3389/fonc.2022.819128 OpenUrl CrossRef PubMed [67]. ↵ Bakker , A.B. , et al : Melanocyte lineage-specific antigen gp100 is recognized by melanoma-derived tumor-infiltrating lymphocytes . The Journal of Experimental Medicine 179 ( 3 ), 1005 – 9 ( 1994 ) doi: 10.1084/jem.179.3.1005 OpenUrl Abstract / FREE Full Text View the discussion thread. Back to top Previous Next Posted March 29, 2025. Download PDF Supplementary Material Data/Code Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following COMIC: Explainable Drug Repurposing via Contrastive Masking for Interpretable Connections Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share COMIC: Explainable Drug Repurposing via Contrastive Masking for Interpretable Connections Naafey Aamer , Muhammad Nabeel Asim , Andreas Dengel bioRxiv 2025.03.26.645428; doi: https://doi.org/10.1101/2025.03.26.645428 Share This Article: Copy Citation Tools COMIC: Explainable Drug Repurposing via Contrastive Masking for Interpretable Connections Naafey Aamer , Muhammad Nabeel Asim , Andreas Dengel bioRxiv 2025.03.26.645428; doi: https://doi.org/10.1101/2025.03.26.645428 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Bioinformatics Subject Areas All Articles Animal Behavior and Cognition (7629) Biochemistry (17660) Bioengineering (13881) Bioinformatics (41911) Biophysics (21436) Cancer Biology (18578) Cell Biology (25482) Clinical Trials (138) Developmental Biology (13371) Ecology (19887) Epidemiology (2067) Evolutionary Biology (24302) Genetics (15599) Genomics (22483) Immunology (17728) Microbiology (40364) Molecular Biology (17163) Neuroscience (88537) Paleontology (666) Pathology (2830) Pharmacology and Toxicology (4821) Physiology (7637) Plant Biology (15129) Scientific Communication and Education (2045) Synthetic Biology (4290) Systems Biology (9817) Zoology (2269)