The new coronavirus belongs to the β genus of the coronavirus family, with SARS and MERS being its close relatives. While several drug repurposing and vaccine development projects are underway, there is currently no effective cure for COVID-19, and the pandemic continues to spread at an alarming rate. In the race to find an effective treatment, the team at XtalPi quickly assembled a task force and used our forte in quantum physics, AI, and cloud-computing technologies to conduct calculations that can help us explore and understand the virus's infection and treatment mechanism on the molecular level. We are sharing our models and results and will continue to update new findings. Hopefully, our predictions, as well as the many studies published by the global scientific community, will contribute to the acceleration of experimental research and expedite the finding of an effective treatment.
1/20: XtalPi assembled a special task force to study the new coronavirus strand, and started to created homology models of the virus’s important protein subunits based on the genome sequence of the virus released by NCBI on the same day, including two potential drug targets identified previous studies on SARS, PLpro and 3CLpro, and the spike protein.
- 1/21: By comparing the genome sequence of the new coronavirus (dubbed SARS-CoV-2) to the SARS virus (SARS-CoV), we created homology models of 3 important protein subunits of SARS-CoV-2: non-structural proteins 3CLpro and PLpro, which play an important role in the transcription and duplication of the virus and have previously been identified as ideal drug targets in SARS and MERS researches, and the binding configuration of the spike glycoprotein’s receptor binding domain (RBD) and human ACE2 receptor.
- 1/23: Based on the homology structure of SARS-CoV-2, we started to scan through over 2,900 FDA approved drug molecule database for molecules that can effectively bind to the 3CLpro and PLpro targets; conducted binding mode analysis of high-profile antiviral drugs that have been reported to inhibit other strands of coronavirus in experiments and other research literature, including Remdesivir. We also performed computational alanine scanning mutagenesis on the "hotspot" residues at Spike protein's RBD and ACE2's binding interfaces using relative free energy calculation to evaluate SARS-CoV-2's ability to infect humans and spread among humans. Our calculation suggests SARS-CoV-2's mutations lead to a greater binding affinity to ACE2 relative to SARS-CoV.
- 1/26: Found 95 FDA approved drugs with relatively strong binding affinity to the 3CLpro and PLpro targets. Started conducting high-accuracy computational chemistry calculations to refine their binding affinity ranking and narrow down the list of candidates.
- 1/27: Started screening through a database of over 10,000 of natural compounds and Chinese traditional medicine molecules for potential inhibitors of the 3CLpro and PLpro targets
- 1/29: Started using free energy perturbation (XFEP) calculation to predict potential mutations on the spike protein and their effect on its binding affinity to ACE2 to identify mutations that could potentially increase SARS-CoV-2's infectious and pathogenic ability.
- 1/30: Found 87 natural compounds in drug repurposing screening.
- If you are interested in obtaining the list of 95 drug molecules from our drug repurposing screening, please email us at firstname.lastname@example.org
- 2/1: Finished preliminary alanine screening of spike protein mutations that could lead to stronger binding affinity to ACE2. We will continue the screening and will translate and publish our findings in English. Read our post on this study in Chinese here
- 2/6: Based on XFEP calculation results, we have further narrowed down the list of drug repurposing candidates to 38.
- 2/14: Purchased proteins and molecules to experimentally validate the potency of our list of drug repurposing candidates. Summarized potential cell models and animal models that can be used for SARS-CoV-2 research. Read the Chinese post here.
- 2/18: Jan 28, it was reported that Chloroquine has shown efficiency in inhibiting SARS-CoV-2 in cell models. Using molecular dynamics simulation, we started to study its possible mechanisms in potentially treating COVID-19. Based on the literature hypotheses, we tested 1 of the 3 suspected mechanisms and verified its possibility, and found a new one - that the molecule could bind to the PLpro target on the virus. We published our result one day after Chloroquine was reported to have demonstrated clear therapeutic effects in treating COVID-19 patients in preliminary clinical trials in China. XtalPi is collaborating with Zhongsheng to test out these findings in experiments. Read our post in Chinese here. (Chroloquine has since been added to China's official COVID-19 treatment guidelines).