Visual Analytics and Information Visualization

How we help you see things more clearly

We at BioSeek believe that the process of sense-making of data in research must be assisted to the greatest extent possible in order to save researchers’ time and energy. Information visualization can help progress considerably. We strive to deliver the clearest representations of large amounts of data in a small space, enhance the recognition of patterns, make large data sets coherent, despite their complexity. And make it all look beautiful.

Main analytical tools on BioSeek


Visualizes connections between entities, based on mentions of these entities in the scientific publications in our database. Entities are visually represented as bubbles, each individual bubble represents one single entity. The larger the bubble, the more mentions this entity has. The article mentions act as gravitational force between the bubbles, therefore clusters are formed by entities which are often co-mentioned in the same article. The clusters are signs of existing or emerging theories. This algorithm is based on T-distributed Stochastic Neighbor Embedding (t-SNE) methodology. More about it in this video In this example clustering is performed around the gene nfkb1. It is interactive and explorable


When your search query contains more than 3 entities, the Heatmap performs searches for articles which mention each possible pair of entities from your original search and delivers results grouped in a table. In this case, our query contains the following entities: “Atopic Dermatitis”, “FLG”, “LOR”, “Psoriasis” and "Mice".

Citations Analytics

When you want to know which entities are being referred to most often, this is your tool. The Citations Analytics visualizes the total number of all mentions of given entity in articles per year. In this case we are searching the gene rplK. Each dot opens the specific article in the Right Panel.

Timeline Analytics

Shows number of articles in which given entity is mentioned, for each year. In this case, we search Atopic Dermatitis. We see, for example, that the peak of mentions of the gene rplK was in 2004.

Stream Analytics

Shows popularity tendencies of entities within given group in time. For example, if you search rplK, and apply this tool, it will visualize the Top 10 genes co-mentioned with rplK for a given period, the default being 10 years.

Timeline- 2-nd layer

Shows number of co-mentions of searched entity with additionally selected second (third, fourth, etc.), entity from lists. In this case: searched entity is nfkb1, additionally selected selected second entity from gene list is relA.

The Insights Section

This one is quite self-explanatory, but we’ll add some detail: user behavior analysis generates statistics on user’s searches, views, interactions; Artificial Intelligence is in charge of selecting and delivering the most relevant info for the specific user in curated personalized lists of suggestions- not only articles, but also people — if they have explicitly allowed this, of course. The Insights section also gives you a detailed and clear overview of your actions, making it extremely easy to keep track of things.

Bringing it all together:

Visual analysis helps humans process data and information faster. Multi-dimensional visual representations are important, automatic data visualization significantly reduces work time and helps people think visually. Shifting perspectives on data is critical to understanding. Linking perspectives across dashboards and views makes analysis faster and more effective. Last, but not least: data visualization naturally extends collaboration across organizations. That’s why great information visualization is a main priority on