Use case 1

Amrun et al. present four different immunodominant B cell assay epitopes, to be used as highly specific and sensitive serological diagnostic targets, i.e., to test for the presence of the virus in patients potentially exposed to SARS-CoV-2. The candidate epitopes are named S14P5, S20P2, S21P2 on the Spike protein, and N4P5 on the N protein. Amrun et al. authors studied the conservation of these epitopes across ~17k SARS-CoV-2 sequences publicly available at the time of their writing. They reported a low rate of potential amino acid changes over the epitopes.

By using EpiViruSurf Mode 1, it is possible to replicate such epitopes as ranges of positions on the proteins, respectively on [553,570], [769,786], [809-826] on Spike and [153-170] on N. These may be checked, for instance, against the EpiViruSurf sequence population from the USA, of about 205k sequence as of May 9th, 2021.

In this figure, we show a particular snapshot of the analysis session where the user has already inserted all four epitopes. See the third red rectangle framing the user-defined epitopes, where -- for brevity -- we only show the card produced for the first S14P5 epitope.

When all four have been inserted, the user will be provided with the results framed by the green rectangle (1). It is worth noting that the first epitope has a high ratio of altered sequences, i.e., 25.1%. The user may be interested in inspecting the breakdown of such a consistent set of sequences.

By clicking on the number of total amino acid changes (i.e. TOT MUT), we open the "Epitope mutation statistics" functionality. We may group by the country attribute, thereby obtaining a table that, for each amino acid change, reports the total of sequences exhibiting such changes, and the breakdown of such amount by 'country'.

Through sorting by descending total count, we observe that the Spike A570D position is the most commonly mutated one. We also check the most impacted US states (grouping by the attribute 'region'), which are Florida, Minnesota, and Michigan. An alternative grouping can be performed on `lineage', highlighting that the sequences with this mutation are almost always assigned to the B.1.1.7 lineage, corresponding to the Variant of Concern (first defined on a Virological.org post in December 2020). We can make our conservancy analysis more specific by adding new epitopes to our list, tested against smaller populations; in the figure, Result section, box (2) we have created the candidate epitopes S14P5-USA-FL, S14P5-USA-MN, and S14P5-USA-MI. In the MUT SEQ RATIO column, we observe that Minnesota (MN) and Michigan (MI) have a higher incidence of mutations on this epitope.

We then focus on N4P5 (reported by Amrun et al. as the most stable epitope out of the four). In our knowledge base (described in the Background section), we find three amino acid changes falling within the scope of this epitope (namely, A156S, L161F, P168S); it has been claimed that they may lower the protein stability and modify the protein flexibility (see Rahman et al., 2021. Due to these changes, there is the possibility that the specificity and sensitivity of serological tests for COVID-19 diagnosis may be impacted (leading to false negatives). In the figure:

• In the first red box, we show how we insert the custom epitope condition
• In the second red box, we input an amino acid condition, N protein, 168-168 position range for a substitution for P (Proline) to S (Serine). By adding this condition we are instructing the system to compute statistics only over the fraction of the selected population (all SARS-CoV-2, human host sequences in this example) that carry the specified amino acid change.
• in the result section, green box (3), we observe that the ratios of the mutated sequences for the N45P5 epitope over the three defined populations are quite low (below 0.5%).

Note that, if the MUT FREQ is close to 1 -- when one of these three changes is present -- no other change is carried within the epitope scope. Overall, the impacts on the epitope are minimal but attention should be paid and further investigations needed to proceed with its use for serologic assays.