EMoMiS: A Pipeline for Epitope-based Molecular Mimicry Search in Protein Structures with Applications to SARS-CoV-2

Article activity feed

1. 

   ScreenIT

   Feb 11, 2022

   SciScore for 10.1101/2022.02.05.479274: (What is this?)

   Please note, not all rigor criteria are appropriate for all manuscripts.

   Table 1: Rigor

   NIH rigor criteria are not applicable to paper type.

   Table 2: Resources

   Antibodies
   Sentences	Resources
   The Molecular Surface Interaction Fingerprint Search (MaSIF-Search) that uses a geometric deep learning approach was used to evaluate antibody-antigen binding, (Gainza et al., 2020).	antibody-antigen binding, suggested: None
   In Step D of the EMoMiS pipeline, we evaluate the structurally similar motifs for the binding strength of the antibody-antigen pair.	antibody-antigen pair. suggested: None
   Software and Algorithms
   Sentences	Resources
   The EMoMiS pipeline uses the three-dimensional structures, SDB as well as sequence information, QDB of antibody-antigen complexes from the structural antibody database SAbDab …

   SciScore for 10.1101/2022.02.05.479274: (What is this?)

   Please note, not all rigor criteria are appropriate for all manuscripts.

   Table 1: Rigor

   NIH rigor criteria are not applicable to paper type.

   Table 2: Resources

   Antibodies
   Sentences	Resources
   The Molecular Surface Interaction Fingerprint Search (MaSIF-Search) that uses a geometric deep learning approach was used to evaluate antibody-antigen binding, (Gainza et al., 2020).	antibody-antigen binding, suggested: None
   In Step D of the EMoMiS pipeline, we evaluate the structurally similar motifs for the binding strength of the antibody-antigen pair.	antibody-antigen pair. suggested: None
   Software and Algorithms
   Sentences	Resources
   The EMoMiS pipeline uses the three-dimensional structures, SDB as well as sequence information, QDB of antibody-antigen complexes from the structural antibody database SAbDab (Dunbar et al., 2014).	EMoMiS suggested: None
   In Step D, we evaluate the potential cross-reactivity using a pre-trained deep learning model (Gainza et al., 2020), which estimates the binding strength between the antibody SDB-Ab, complexed with the database antigen SDB and the target antigen Starget-Ag.. 2.1. Sequence Similarity Search: The sequence similarity search between the target protein sequence, Qtarget, and the antigen sequences from the data set, QDB, was performed using Protein-Protein BLAST 2.12.0+ (Altschul, 1997).	Protein-Protein suggested: (Protein-protein interfaces, RRID:SCR_007879) BLAST suggested: (BLASTX, RRID:SCR_001653)
   FASTA sequences were downloaded using BioPython utilities (Cock et al., 2009).	BioPython suggested: (Biopython, RRID:SCR_007173)
   For 3000 random antigens from the SAbDab database, we isolated possible epitopes of lengths ranging from 3 to 32 AA, such that the center of each motif was in contact with its native antibody.	SAbDab suggested: None
   To ensure reproducibility, the running environment was containerized with Singularity version 3.8.5 (Kurtzer et al., 2017).	Singularity suggested: (Singularity Hub, RRID:SCR_016248)

   ---

   Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

   ---

   Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:

   Another limitation of the deep learning model is the absence of glycans in the set of features. The glycosylation events may significantly affect the antibody neutralizing properties and thus, the model sensitivity (Miranda et al., 2007). Those limiting factors may consolidate the false-negative predictions. For example, the score for the antibody from the Ab-Spike OC43 complex (PDB ID 7M51) to bind SARS-CoV-2 Spike protein (PDB ID 7M53) was very low (DL score = 3.066, p-value = 0.306, see Supplemental Table S3). Yet, the cross-reaction between antibody B6 (PDB IDs 7M51 and 7M53) with coronavirus OC43 and SARS-CoV-2 Spike proteins was experimentally verified (Sauer et al., 2021). On the other hand, another Spike configuration (PDB ID 7RNJ) was found to cross-react with coronavirus OC43 (Table 1, H), which highlights the advantage of using multiple target structures. While the lack of a molecular structure for the target protein may be seen as a limitation for the EMoMiS pipeline, protein structures can be quickly and accurately predicted with the recent advances in the deep learning field (Jumper et al., 2021). 4.2. Antiviral Antibody Cross-reaction: The EMoMiS pipeline identified four sites in the SARS-CoV-2 Spike protein that may mimic epitopes from other viruses. Antibody cross-reaction with viral mimic epitopes may provide cross-protection to the host. On the other hand, antibodies elicited by mimicry may cross-react with the vital antigen with lower affinity. As a result...

   ---

   Results from TrialIdentifier: No clinical trial numbers were referenced.

   ---

   Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

   ---

   Results from JetFighter: We did not find any issues relating to colormaps.

   ---

   Results from rtransparent:

   • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
   • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
   • No protocol registration statement was detected.

   ---

   Results from scite Reference Check: We found no unreliable references.

   ---

   Read the original source
2. 

   Version published to 10.1101/2022.02.05.479274 on bioRxiv

   Feb 7, 2022