ConTour: Data-Driven Exploration of Multi-Relational Datasets for Drug Discovery

Abstract

Large scale data analysis is nowadays a crucial part of drug discovery. Biologists and chemists need to quickly explore and evaluate potentially effective yet safe compounds based on many datasets that are in relationship with each other. However, there is a lack of tools that support them in these processes. To remedy this, we developed ConTour, an interactive visual analytics technique that enables the exploration of these complex, multi-relational datasets. At its core ConTour lists all items of each dataset in a column. Relationships between the columns are revealed through interaction: selecting one or multiple items in one column highlights and re-sorts the items in other columns. Filters based on relationships enable drilling down into the large data space. To identify interesting items in the first place, ConTour employs advanced sorting strategies, including strategies based on connectivity strength and uniqueness, as well as sorting based on item attributes. ConTour also introduces interactive nesting of columns, a powerful method to show the related items of a child column for each item in the parent column. Within the columns, ConTour shows rich attribute data about the items as well as information about the connection strengths to other datasets. Finally, ConTour provides a number of detail views, which can show items from multiple datasets and their associated data at the same time. We demonstrate the utility of our system in case studies conducted with a team of chemical biologists, who investigate the effects of chemical compounds on cells and need to understand the underlying mechanisms.

The State of the Art in Visualizing Multivariate Networks

“…A prominent example is Semantic Substrates [SA06], Figure (a), which facet a network based on a categorical attribute and let analysts choose whether to show links between or within facets, or both. This approach is also commonly used in biological networks to visualize nodes within spatially segregated cell compartments [BMGK08], or to lay out k‐partite or multityped graphs [SJUS08, PLS*14, PvW08, GKL*13, ABF*07]. There are alternatives to faceting by sets or node types: Figure (b) shows faceting based on a hierarchical clustering algorithm where the clusters are laid out in a treemap [RMS*11] and nodes are shown within the treemap cells.…”

“…Semantic Substrates [SA06, AS07], for example, place nodes according to other attribute values, whereas Cerebral [BMGK08] uses a layout optimized for topology. A series of other technique use a linear layout [PvW08, SJUS08, PLS*14], which is amenable to attribute visualization.…”

MediSyn: uncertainty-aware visualization of multiple biomedical datasets to support drug treatment selection

“…Several research efforts visualize datasets in separate views and then use linking and brushing techniques or explicit links to show data relations. ConTour [ 22 ] provides a relationship view of datasets, such as genes, compounds, and pathways, in columns at the bottom with a detailed view of the selected items above. The user selection of items in one column can highlight relevant items in other columns.…”

Progress in visual representations of chemical space

“…Nevertheless, in the course of the analysis of the structure--activity relationships (SAR) in a series of closely related analogs, it is important to compare them not only by gross measures, but also by pairwise relations. The network and, more generally, graph analysis is widely used in computational and mathematical chemistry for a long time and was applied to numerous tasks, such as classification of chemical reactions [75], enumeration of small molecules [76], data organization [77], and QSAR model generation [78]. Drug discovery-related networks in the paradigm of network pharmacology [79] consider compound--target pairs (e.g., [80]) as well as target--target pairs (e.g., [81]); much less attention is paid to compound--compound relations [82].…”