We build a nanopublication-based knowledge graph of protein/protein, drug/protein, and protein/disease interactions to create a resource for exploring potential therapies for diseases. This is accomplished using Semantic Web standard tools like Blazegraph, SADI web services, and JSON-LD for integration with a Javascript-based web client. Metastatic cutaneous melanoma is an aggressive skin cancer with some progression-slowing treatments but no known cure. The omics data explosion has created many possible drug candidates, however ltering criteria remain challenging, and systems biology approaches have become fragmented with many disconnected databases. Using drug, protein, and disease interactions, we built an evidence-weighted knowledge graph of integrated interactions. Our knowledge graph-based system, ReDrugS, can be used via an API or web interface, and has generated 25 high quality melanoma drug candidates. We show that probabilistic analysis of systems biology graphs increases drug candidate quality compared to non-probabilistic methods. Four of the 25 candidates are novel therapies, three of which have been tested with other cancers. All other candidates have current or completed clinical trials, or have been studied in in vivo or in vitro. This approach can be used to identify candidate therapies for use in research or personalized medicine.
Metastatic cutaneous melanoma is an aggressive cancer of the skin with low
prevalence but very high mortality rate, with an estimated 5 year survival
rate of 6 percent [
A knowledge graph is a compilation of facts and gures that can be used to
provide contextual meaning to searches. Google is using knowledge graphs to
improve its search and to analyze the information graph of the web; Facebook is
using them to analyze the social graph. We built our knowledge graph with the
goal of unifying large parts of biomedical domain knowledge for both mining and
interactive exploration related to drugs, diseases, and proteins. Our knowledge
graph is enhanced by the provenance of each fragment of knowledge captured,
which is used to compute the con dence probabilities for each of those fragments.
Further, we use open standards from the World Wide Web Consortium (W3C),
including the Resource Description Framework (RDF) [
We built a novel computational drug repositioning platform, that we refer to as ReDrugS, that applies probabilistic ltering over individually-supported assertions drawn from multiple databases pertaining to systems biology, pharmacology, disease association, and gene expression data. We use our platform to identify novel and known drugs for melanoma. 2
We used ReDrugS to examine the drug-target-disease network and identify known, novel, and well supported melanoma drugs. The ReDrugS knowledge base contained 6,180 drugs, 3,820 diseases, 69,279 proteins, and 899,198 interactions.
We examined drug and gene connections that were 3 or less interaction steps from melanoma, and additionally ltered interactions with a joint probability greater or equal to 0.93. We identi ed 25 drugs in the resulting drug-gene-disease network surrounding melanoma as illustrated in Figure 1 .
We then validated the set of 25 drugs by determining their position in the drug discovery pipeline for melanoma. Nearly all drugs uncovered by ReDrugS were previously been identi ed as potential melanoma therapies either in clinical trials or in vivo or in vitro. Of the 25 drugs, 12 have been in Phase I, II, or III clinical trials, 5 have been studied in vitro, 4 in vivo, 1 was investigated as a case study, and 3 are novel.
To further evaluate our system, we examined the impact of decreasing the joint probability or increasing the number of interaction steps. Figures 2 A and B show precision, recall, and f-measure curves while varying each parameter. Using these information retrieval performance curves we found that using a joint probability of 0.93 or greater with 3 or less interaction steps maximizes the precision and recall as shown in Figure 2.
By performing a literature search on hypothesis candidates with a joint probability of 0.5 or higher and 6 or fewer interaction steps, we were able to generate precision, recall, and f-measure curves for both cuto s to nd our cuto of 0.93 with 3 or fewer interaction steps. The precision, recall, and f-measure curves are shown for varying joint probability thresholds in Figure 2 A and for varying interaction step counts in Figure 2 B. 3
We designed ReDrugS to quickly and automatically integrate and lter a heterogeneous biomedical knowledge graph to generate high-con dence drug repositioning candidates. Our results indicate that ReDrugs generates clinically plau(A) Information Retrieval by Probability Threshold precision recal fmeasure (B) Information Retrieval by Network Expansion Step precision recal fmeasure 0.75 0.5 0.25 sible drug candidates, in which half are in various stages of clinical trials, while others are novel or are being investigated in pre-clinical studies. By helping to consolidate the three main datatypes - drug targets, protein interactions, and disease genes, ReDrugs can amplify the ability of researchers to lter the vast amount of information into those that are relevant for drug discovery. 3.1
ReDrugS uses a fairly straightforward web architecture, as shown in Figure 3. It uses the Blazegraph RDF database backend. The database layer is interchangeable except that the full text search service needs to use Blazegraph-only properties to perform text searches as text indexing is not yet standardized in the SPARQL query language. All other aspects are standardized and should work with other RDF databases without modi cation. ReDrugs currently uses the Python-based TurboGears web application framework hosted using the Web Services Gateway Interface (WSGI) standard via an Apache HTTP server. TurboGears in turn hosts the SADI web services that drive the application and access the database. It also serves up the static HTML and supporting les.
The user interface is implemented with AngularJS and Cytoscape.js, which submits queries to the SADI web services using JSON-LD and aggregates results into the networked view. The software relies exclusively on standardized protocols (HTTP, SADI, SPARQL, RDF, and others) to make it simple to replace technologies as needed. The data itself is processed using conversion scripts as shown in Figure 4.
Javascript Web Client
Cytoscape.js /api/search /api/upstream /api/downstream Python + Apache Web Server
RDF Store
This research project did not involve human subjects. The ReDrugS platform
consists of a graphical web application, an application programming interface
(API), and a knowledge base. The graphical web application enables users to
initiate a search using drug, gene, and disease names and synonyms. Users can
then interact with the application to expand the network at an arbitrary number
of interactions away from the entity of interest, and to lter the network based on
a joint probability between the source and target entities. Drug-protein,
proteinprotein, and gene-disease interactions were obtained from several datasets and
integrated into ontology-annotated and provenance and evidence bearing
representations called nanopublications. The web application obtains information
from the knowledge base using semantic web services. Finally, we evaluated
our approach by examining the mechanistic plausibility of the drug in having
melanoma-speci c disease modifying ability. We evaluated a large number of
possible drug/disease associations with varying joint probabilities and interaction
steps to determine the thresholds with the highest F-Measure, resulting in our
thresholds of three or less interactions and a joint probability of 0.93 or higher.
We developed four Semantic Automated Discovery and Integration (SADI) web
services [
Ontological Resources Protein/Protein Interaction Ontology, Semanticscience Integrated Ontology,
Gene Ontology vocabularies, relationships
ReDrugS RDF Store queries evidence to probability
Experimental
Method
Assessment Confidence scores of experimental methods. iRefIndex converted to nanopubs
ReDrugS API Interaction network search
and expansion queries graph
queries graph
Analytical Tools Cytoscape, R, Python, etc.
ReDrugS Cytoscape.js App
The rst service is a simple free text lookup, that takes an pml:Query 5 [
The other three SADI services look up interactions that contain a named entity. Two of them look at the entity to nd upstream and downstream connections, and the third service assumes that the entity is a biological process and nds all interactions that related to that process. The services return only one interaction for each triple (source, interaction type, target). There are often multiple probabilities per interaction, and more than one interaction per interaction type. This is because the interaction may have been recorded in multiple databases, based on di erent experimental methods. To provide a single probability score for each interaction of a source and target, the interactions are combined. A single probability is generated per identi ed interaction by taking the geometric mean of the probabilities for that interaction. However, this 5 PML 3, in development: https://github.com/timrdf/pml. This includes PML 2 constructs that are not covered in PROV-O. Service Description URL Input Output Name Resource Look up resources using search pml:Query text tearch free text search against their RDFS labels. This service is optimized for typeahead user interfaces.
Find inter- Find interactions whose process sio:Process actions in participants or targets also a biological participate in the input process process.
Find up- Find interactions that the upstream stream input entity is a target of participants in and have explicit partic
ipants.
Find down- Find interactions that the downstream sio:MaterialEntity sio:Agent stream tar- input entity participates in gets and have explicit targets. sio:MaterialEntity sio:Target sio:Process method is undesirable when combining multiple interaction records of the same type. We instead combine the interaction records using a form of probabilistic voting using composite Z-Scores. This is done to model that multiple experiments that produce the same results reinforce each other, and should therefore give a higher overall probability than would be indicated by taking their mean or even by Bayes Theorem. We do this by converting each probability into a Z Score (aka Standard Score) using the Quantile Function (Q()), summing the values, and applying the Cumulative Distribution Function (CDF ()) to compute the corresponding probability:
P (x1:::n) = CDF n X Q (P (xi)) i=1 !
These composite Z Scores, which we transform back into probabilities, are
frequently used to combine multiple indicators of the same underlying
phenomena, as in [
The user interface was developed using the above SADI web services and uses Cytoscape.js,6, angular.js,7 and Bootstrap 3.8 An example network is shown 6 http://cytoscape.github.io/cytoscape.js 7 https://angularjs.org 8 http://getbootstrap.com in Figure 1 Users can search for biological entities and processes, which can then be autocompleted to speci c entities that are in the ReDrugS graph. Users can then add those entities and processes to the displayed graph and retrieve upstream and downstream connections and link out to more details for every entity. Cytoscape.js is used as the main rendering and network visualization tool, and provides node and edge rendering, layout, and network analysis capabilities, and has been integrated into a customized rich web client.
In order to evaluate this knowledge graph, we developed a demonstration web interface9 based on the Cytoscape.js10 JavaScript library. The interface lets users enter biological entity names. As the user types, the text is resolved to a list of entities. The user nishes by selecting from the list, and submitting the search. The search returns interactions and nodes associated with the entity selected, which are added to the Cytoscape.js graph. Users are also able to select nodes and populate upstream or downstream connections. Figure 1 is an example output of this process. 9 http://redrugs.tw.rpi.edu 10 http://cytoscape.github.io/cytoscape.js