In silico cyclic-minipeptides predicting antibody-free early-diagnosis of Ebola virus

In this work, in silico cyclic-minipeptide candidates were computationally generated aiming to develop antibody-free sandwichbased early-diagnosis of ebola virus (EBOV). Presently there are few approved early-rapid EBOV diagnosis, despite many antibody-based efforts. Cyclic-minipeptides were chosen here because i) their vast sequence numbers facilitate designs fitting the diagnostic sensibility and specificity requirements, ii) they are easy and cheaper to scale-up for production, and iii) their rigidity could be more resistant to denaturing-buffers for EBOV early disinfection during assay. In this work, the earlier most abundant biomarker for EBOV infections soluble glycoprotein sGP were computationally targeted. Hundreds of ~ 100-1000-fold 9- and 12-mer candidates predicting higher affinities than previously reported antibodies were proposed here combining Boltzgen/ Prodigy with our newly developed evolutionary-mutations. By targeting complementary sGP cavities with cyclic-minipeptide candidates, possible sandwich-pairs would still require experimentally-guided explorations. Additional computational screening work may also be possible to select the best candidates for possible pan-EBOV diagnosis.