top of page



April 20, 2020

As society attempts to get ahead of the spread of the 2019 novel coronavirus (which causes the disease COVID-19), in the drug discovery world that means an unprecedented race to find new treatment protocols. Now more than ever, when days and weeks of advancement can make all the difference in mitigating this pandemic, it’s time to look to augmented intelligence (AI) tools that can accelerate the human know-how already at work.

To this end, we’re launching a series aimed at understanding the underlying biological mechanisms at work using our AI knowledge mapping tool. (We’re also offering our tool to coronavirus researchers at no cost. Contact us here for more information.)

Using inputs ranging from a single search term to multiomics datasets, our AI tool assembles information from multiple knowledge sources, removes the highly redundant data, draws conclusions about the interconnected data, and displays a holistic, interactive map. Researchers gain more insights than they would uncover by simply searching knowledge sources on their own, and in a much shorter amount of time.

Given the recent interest in hydroxychloroquine/chloroquine as a potential drug for the treatment and prevention of COVID-19, our team sought to investigate the mechanisms by which chloroquine may or may not be an effective drug in this context. In the following sections, we will present the information our AI tool found about the coronavirus, chloroquine, and potentially important biological interactions between these two queries.


By exploring our AI-generated knowledge maps for each search, our team was able to rapidly get up to speed on the biology surrounding the coronavirus and chloroquine, and also identify both leading and emerging hypotheses for how chloroquine may target the coronavirus.


NOTE: We are not recommending or endorsing this treatment protocol; this is simply a case study in exploring biological mechanisms through our AI tool.


To start, our team submitted the search term “coronavirus” into our AI tool. Below, you’ll see the resulting knowledge map. Most of the biological themes (shown as spheres) were consistent with the scientific community’s basic understanding of the coronavirus. For example, in pink, you’ll see themes illustrating the respiratory sequelae stemming from infection. By clicking on a theme, you are then able to view the provenance of the insight (e.g., the abstract of the article), all within the visualization platform.

Coronavirus (1).png

General overview of themes related to "coronavirus":

  • Viral replication and virulence themes are in red.

  • Themes related to cleavage/activation of virus for host cell entry are in blue.

  • Themes containing antiviral targets are in orange.

  • Mucosal and enteric responses to the virus are in purple themes.

  • Themes showing respiratory sequelae from infection are in pink.

  • Host response to Coronavirus is depicted in teal themes.


Key takeaways:

Next, we wanted to gain insight into the mechanisms of chloroquine. For this section we ran a search for “chloroquine,” a quinoline compound with the same core mechanism of action as hydroxychloroquine, but with a more robust scientific publication history. The concept map below shows the interconnected nature of chloroquine mechanisms, such as the disruption of the endo-lysosome system, which thereby blocks the pH-dependent endocytosis viral entry pathway. Our findings are consistent with the main takeaways of a recent review article in Nature summarizing the mechanisms of chloroquine/hydroxychloroquine.

Chloroquine (1).png

Chloroquine Map:

  • Autophagy-related themes are in blue. Chloroquine is a widely used autophagy inhibitor due to its ability to inhibit autophagosome-lysosome fusion. 

  • Endocytosis-related themes are in red. Chloroquine impairs the endo-lysosomal system.

  • Toll-like receptor (TLR) signaling is in yellow.

  • Infectious disease-related themes are in purple. Most of these are related to malaria, as chloroquine and hydroxychloroquine are widely used for malaria prophylaxis and treatment.

  • Other therapeutic uses for chloroquine, besides preventing infectious diseases, are in pink.

  • Adverse effects or contraindications for taking chloroquine are in green.

Key takeaways:

  • Chloroquine is widely used to prevent infectious diseases such as malaria, but is also used for other indications such as rheumatoid arthritis.

  • Chloroquine use comes with risks; it may be associated with arrhythmia, retinopathy, pruritis, and other adverse effects.

  • The most prominent suggested mechanism of action of chloroquine is the impairment of the endo-lysosomal system, which is the viral entry pathway for the SARS-coronavirus.


Finally, to examine the interaction of coronavirus and chloroquine, we used the filter feature in CompBio to overlay the two and find common links. Here is the resulting concept map, which highlights themes such as lysosomes as possible mechanisms for how chloroquine may prevent coronavirus infection. Additionally, research suggests that SARS-CoV-2 disrupts heme in erythrocytes. That suggests chloroquine may prevent the virus from attacking heme.

Interaction between Coronavirus and Chloroquine:


Key takeaways:

  •  The most prominent conserved themes are related to the endo-lysosome system, indicating the likely mechanism by which chloroquine could prevent Coronavirus infection.

  • Other interesting themes include “Erythrocytes” and “TLR signaling.”

    • Some research suggests that SARS-CoV-2 disrupts heme in erythrocytes, and that chloroquine may prevent the virus from attacking heme.

    • Chloroquine has also been suggested to inhibit inflammatory cytokine production by inhibiting endosomal TLRs (summarized here).

To view a video of the visualizer for the overlap of chloroquine and coronavirus, click here


In conclusion, our AI tool allowed us to quickly convert available research findings into an easy-to-understand visualization, which helped our computational biologists make connections of their own. The ability for researchers to become well-versed in the current literature and identify scientifically grounded, testable hypotheses in a timely manner will accelerate the timeline to effective drug treatments.


If you have thoughts on this analysis, or if you have COVID-19 data you'd like us to run, free of charge, send us a messageso our team can contact you with potential next steps.

This Venn diagram illustrates how the concept filter function compares the biology conserved between two search terms. Click the monitor icon below to see the video of conserved biology.


Share your impressions

Send us an idea or share your feedback below. We may test it in CompBio, and create and share hypotheses on the blog.*

*We are launching this webpage to share knowledge maps of Coronavirus and to encourage hypotheses generation. We invite you to share your ideas, insights, and feedback (“Feedback”) to help us create hypotheses in connection with Coronavirus therapies, treatments, and solutions. Please click on the links below which cover our Right to Use Feedback, and important Disclaimers. By submitting Feedback, you agree that you have read and agree to these terms. Thank you for your participation in this project.

bottom of page