docs(manuscript): minor textual refinements

Author: Mnikley

manuscript/biorxiv/main.tex

@@ -62,7 +62,7 @@
 
 % ── Abstract ─────────────────────────────────────────────────────────────────
 \begin{abstract}
-\noindent\textbf{Motivation:} Biological network visualization together with graph-based analyses are key techniques in systems biology and network medicine to detect patterns and generate new hypotheses regarding disease pathobiology, drug target identification, or digital drug discovery. Network representations are also a way to communicate research findings and share results with colleagues and coworkers. \\
+\noindent\textbf{Motivation:} Biological network visualization together with graph-based analyses are key techniques in systems biology and network medicine to detect patterns and generate new hypotheses regarding disease pathobiology, drug target identification, biomarker prioritization, or digital drug discovery. Network representations are also a way to communicate research findings and share results with colleagues and coworkers. \\
 \textbf{Results:} We have developed Graph Lens Lite, a browser-based tool that combines rich visualization capabilities with a streamlined interface for exploring and sharing biological networks. It offers an expressive query language, topological network analysis, GUI-based filtering, visual grouping, customizable layouts, a data-editor, and fine-grained property-based styling options, particularly suited for visualizing molecular models of disease pathobiology or drug mechanism of action. \\
 \textbf{Availability:} Graph Lens Lite is available at GitHub (\url{https://github.com/Delta4AI/GraphLensLite}). \\
 \textbf{Supplementary information:} Supplementary data are available at \textit{Bioinformatics} online.

manuscript/oxford/main.tex

@@ -60,7 +60,7 @@
 \address[2]{\orgdiv{Division of Pediatric Nephrology and Gastroenterology, Department of Pediatrics and Adolescent Medicine, Comprehensive Center for Pediatrics}, \orgname{Medical University Vienna}, \orgaddress{\street{Waehringer Guertel 18-20}, \postcode{1090}, \state{Vienna}, \country{Austria}}}
 \address[3]{\orgdiv{Department of Internal Medicine IV}, \orgname{Medical University Innsbruck}, \orgaddress{\street{Anichstrasse 35}, \postcode{6020}, \state{Innsbruck}, \country{Austria}}}
 
-\corresp[$\ast$]{Corresponding authors. \href{mailto:matthias.ley@delta4.ai}{matthias.ley@delta4.ai}, \href{mailto:klaus.kratochwill@delta4.ai}{klaus.kratochwill@delta4.ai}\\}
+\corresp[$\ast$]{Corresponding authors: \href{mailto:matthias.ley@delta4.ai}{matthias.ley@delta4.ai}, \href{mailto:klaus.kratochwill@delta4.ai}{klaus.kratochwill@delta4.ai}\\}
 
 \corresp[$\#$]{Contributed equally}
 
@@ -70,7 +70,7 @@
 
 % ── Abstract ─────────────────────────────────────────────────────────────────
 \abstract{\mdseries
-\textbf{Motivation:} Biological network visualization together with graph-based analyses are key techniques in systems biology and network medicine to detect patterns and generate new hypotheses regarding disease pathobiology, drug target identification, or digital drug discovery. Network representations are also a way to communicate research findings and share results with colleagues and coworkers. \\
+\textbf{Motivation:} Biological network visualization together with graph-based analyses are key techniques in systems biology and network medicine to detect patterns and generate new hypotheses regarding disease pathobiology, drug target identification, biomarker prioritization, or digital drug discovery. Network representations are also a way to communicate research findings and share results with colleagues and coworkers. \\
 \textbf{Results:} We have developed Graph Lens Lite, a browser-based tool that combines rich visualization capabilities with a streamlined interface for exploring and sharing biological networks. It offers an expressive query language, topological network analysis, GUI-based filtering, visual grouping, customizable layouts, a data-editor, and fine-grained property-based styling options, particularly suited for visualizing molecular models of disease pathobiology or drug mechanism of action. \\
 \textbf{Availability:} Graph Lens Lite is available at GitHub (\url{https://github.com/Delta4AI/GraphLensLite}).\\
 \textbf{Contact:} \href{mailto:matthias.ley@delta4.ai}{matthias.ley@delta4.ai}\\

manuscript/shared/content.tex

@@ -23,7 +23,7 @@ \subsection{Installation}
 GLL is a free, open-source web application for visualization and exploration of networks. It is written using HTML, CSS, and JavaScript and runs on any modern browser without requiring additional installations. GLL is distributed under the MIT license and its source code is publicly available on GitHub.
 
 \subsection{Data input}
-GLL accepts spreadsheets containing node and edge tables, or a JSON format preserving the application state. Input data require an `ID' column for nodes and `Source ID/Target ID' columns for edges. Optional columns may specify visual attributes such as labels, shapes, sizes, colors, or coordinates. Custom properties containing user data can be added as new columns, with optional group labels in square brackets. A template with column specifications, example values, and supported data types is available on GitHub and can be generated from within the application. Additionally, a demo loader enables direct fetching of protein-protein interaction networks from the STRING database \citep{szklarczykSTRINGDatabase20252025}, with plans to extend support to further sources. To support onboarding, the application includes an interactive guided tour that walks users through the interface and its core functionalities.
+GLL accepts spreadsheets containing node and edge tables, or a JSON format preserving the application state. Input data require an `ID' column for nodes and `Source ID/Target ID' columns for edges. Optional columns may specify visual attributes such as labels, shapes, sizes, colors, or coordinates. Custom properties containing user data can be added as new columns, with optional group labels in square brackets. A template with column specifications, example values, and supported data types is available on GitHub and can also be generated from within the application. Additionally, a demo loader enables direct fetching of protein-protein interaction networks from the STRING database \citep{szklarczykSTRINGDatabase20252025}, with plans to extend support to further sources. The application includes an interactive guided tour that walks users through the interface and its core functionalities.
 
 \subsection{The graphical user interface}
 The main GUI elements are shown in Figure 1.
@@ -32,7 +32,7 @@ \subsection{The graphical user interface}
 \textbf{(1) Main network view:} The central window for displaying and interactively exploring the loaded network. \\
 \textbf{(2) File import:} Load files in spreadsheet or GLL JSON format. \\
 \textbf{(3) Workspace management:} Save and switch workspaces, reset layout, fit graph to view, and hide disconnected nodes. \\
-\textbf{(4) Data editor:} Interactive table for modifying (adding, editing, deleting) nodes and edges and their properties as well as exporting currently filtered nodes and edges. \\
+\textbf{(4) Data editor:} Interactive table for modifying (adding, deleting, editing) nodes and edges and their properties as well as exporting currently filtered nodes and edges. \\
 \textbf{(5) Query editor:} Custom query language for nested graph filtering supporting Boolean (AND/OR/NOT), comparison (>=,<=,..), and set-membership (IN [..]) operators. \\
 \textbf{(6) Network and image export:} Save the application state to a portable GLL JSON file or export the current view as PNG. \\
 \textbf{(7) Node and edge filtering:} GUI-based filtering and visual grouping, synchronized with the query language. \\
@@ -50,7 +50,7 @@ \subsubsection{Exploring the network}
 Once loaded, users can explore the network structure through multiple approaches. The selection panel (Figure 1.10) enables element search, selection, and expansion to connected neighbors. Metadata for selected node and edge elements appear in specific tooltips (Figure 1.8). For deeper analysis, GLL computes topological metrics (e.g., node centrality measures including degree, betweenness, or PageRank) to identify key nodes within the network (Figure 1.9). Users can select a metric and examine one or more highly ranked nodes, with graph-level statistics displayed in the metric panel. In-app documentation describes each metric's methodology and provides references to relevant literature.
 
 \subsubsection{Filtering the network}
-GLL supports network filtering through graphical controls (Figure 1.7), with an edit mode for precise input (Figure 1.12), or a query editor for complex operations (Figure 1.5). Node and edge properties can be used for filtering and selection through either interface. The query editor uses a left-to-right syntax and ssupports Boolean operators (AND, OR, NOT), comparison (>=,<=), and set membership (IN) operators, as well as nested conditions using parentheses. Contextual help is accessible via an adjacent icon.
+GLL supports network filtering through graphical controls (Figure 1.7), with an edit mode for precise input (Figure 1.12), or a query editor for complex operations (Figure 1.5). Node and edge properties can be used for filtering and selecting through either interface. The query editor uses a left-to-right syntax and ssupports Boolean operators (AND, OR, NOT), comparison (>=,<=), and set membership (IN) operators, as well as nested conditions using parentheses. Contextual help is accessible via an adjacent icon.
 
 \subsubsection{Styling the network}
 The styling panel (Figure 1.11) offers extensive visual customization options for nodes, edges, and groups.  Properties such as geometry (shape, size), color (fill, border, line), labels (text, placement, font), and annotations (badges, halos, arrows) can be configured independently for each element type. Up to four bubble sets allow visual grouping of nodes with adjustable appearance. Numeric and categorical attributes, whether user-supplied or computed from network metrics, can be dynamically mapped to visual features like color and size to highlight importance.
@@ -67,9 +67,9 @@ \subsection{Visualizing a network-based model of autosomal dominant polycystic k
 
 The final constructed ADPKD molecular model consisted of 263 molecular features that were connected by 1574 edges (\textbf{Supplementary datafile S1}). Edges in the constructed network consisted of direct protein-protein interactions consolidated from IntAct \citep{orchardMIntActProjectIntAct2014}, BioGRID \citep{oughtredBioGRIDDatabaseComprehensive2021}, and Reactome \citep{milacicReactomePathwayKnowledgebase2024} as well as of literature sentence-level co-annotations of publications focusing on ADPKD using the catalogs of molecular features and phenotypes as given in Delta4's Hyper-C software platform.
 
-The constructed molecular model can now be used to explore the pathobiology of ADPKD by either filtering for (i) certain molecular mechanisms, (ii) individual cell types, (iii) the most up- or down-regulated genes in renal tissue or specific cell types, (iv) the most central genes based on graph properties like node degree or betweenness among others. Figure 1 for example shows genes of the Wnt signaling pathway (in green for positive modulation of the pathway, red for negative modulation of the pathway, dark blue for modulation of the pathway, and light blue for genes being pathway members without any further information on regulation) as well as additional molecules being associated with ADPKD that are linked to members of the Wnt signaling pathway (in grey) via either protein-protein interactions or co-annotations in the context of ADPKD. This network view allows investigating interesting novel connections of Wnt signaling pathway members with ADPKD-associated proteins.
+The constructed molecular model can now be used to explore the pathobiology of ADPKD by either filtering for (i) certain molecular mechanisms, (ii) individual cell types, (iii) the most up- or down-regulated genes in renal tissue or specific cell types, (iv) the most central genes based on graph properties like node degree or betweenness among others. Figure 1 for example shows genes of the Wnt signaling pathway (in green for positive modulation of the pathway, red for negative modulation of the pathway, dark blue for modulation of the pathway, and light blue for genes being pathway members without any further information on regulation) as well as additional molecules being associated with ADPKD that are linked to members of the Wnt signaling pathway (in grey) via either protein-protein interactions or co-annotations in the context of ADPKD. This network view allows investigating novel connections of Wnt signaling pathway members with ADPKD-associated proteins.
 
-The molecular model can subsequently also be used to (i) screen for compounds showing beneficial impact on dysregulated disease mechanisms as shown previously in other renal diseases \citep{gebeshuberComputationalDrugRepositioning2023}, (ii) identify drug targets of interest, investigate cell-cell interaction analysis to identify relevant receptor-ligand pairs following previously reported analyses \citep{evgeniouMetaAnalysisHumanTranscriptomics2021}, or (iii) prioritize drug targets for the development of new chemical entities.
+The molecular model can subsequently be used to (i) screen for compounds showing beneficial impact on dysregulated disease mechanisms as shown previously in other renal diseases \citep{gebeshuberComputationalDrugRepositioning2023}, (ii) identify drug targets of interest, investigate cell-cell interaction analysis to identify relevant receptor-ligand pairs following previously reported analyses \citep{evgeniouMetaAnalysisHumanTranscriptomics2021}, or (iii) prioritize drug targets for the development of new chemical entities.
 
 \section{Conclusion}
 We have developed Graph Lens Lite for visualizing, exploring and sharing biological networks. GLL is designed for portability with a focus on extensive visual customization. The entire application, together with an average sized graph containing approximately 1000 nodes, can be packed into an archive of only a few megabytes. Unlike comparable tools that may require installation, GLL requires only a web browser for core functionality. GLL is distributed as platform-specific Electron bundles for Windows, macOS, and Linux, as well as a platform-independent, self-contained HTML file. This minimal distribution enables users to load demonstration data or create and work with templates without external dependencies or API calls, ensuring long-term stability and offline operation.