Table 5 List of studies associated with heightened characterization of Abs based on deep sequencing big data assessments
Study | Findings | Limitations |
|---|---|---|
Yang et al. [52] Gallo et al. [53] Barreto et al. [54] | The deep sequencing of in vitro Ab selections enabled the rapid discovery of highly diversified and rare positive clones | ▪ The quality of computational analysis is dependent on target antigen, sample preparation, and background noise removal |
Kelil et al. [55] | The introduction of a motif-based scoring algorithm was used for analyzing deep sequencing datasets derived from in situ Ab selections against cell surface receptors. It enabled effective mining of low-frequency Ab clones (i.e. rare clones) hidden in big data | ▪ The quality of computational analysis is dependent on target antigen, sample preparation, and background noise removal ▪ Sufficient amount of data is required for performing motif-based statistical computations |
Tsioris et al. [57] | A humoral response in infected subjects with West Nile virus is monitored using single-cell BCR deep sequencing analysis. This approach led to the rapid identification of viral-specific Ab clones | ▪ Focuses on natural Abs ▪ Requires infected subjects to derive antigen-selective Abs ▪ Time and resource intensive |
Pan et al. [62] | Isolated bone marrow from immunized mice was subjected to deep sequencing analysis. The introduction of advanced computational screenings helped identify antigen-specific clonal lineages, including Ab frequencies, where these helped determine lead candidate Abs | â–ª Focuses on natural Abs â–ª Requires infected subjects to derive antigen-selective Abs â–ª Time and resource intensive |
Forsyth et al. [66] | A deep sequencing-based CDR mutagenesis scanning platform was used to comprehensively understand clonal sequence affinities. It allowed for the rational design of synthetic Ab sub-libraries and for performing mutagenesis improvements of specific Ab variants | ▪ The point mutations are dependent on experimentally pre-determined structural and mutational data ▪ It is low-throughput and based on time intensive in vitro selections, due to Ab library being displayed on mammalian cells and sorted via flow cytometry |
Friedensohn et al. [27] | A deep-learning model trained on deep sequencing Ab data aided in the effective discovery of convergent features from distinct immune responses. The convergence of Ab features could be determined even if analyzed from distinct immunizations and antigens conditions | ▪ The deep sequencing training datasets are restricted to Ab repertoires of immunized animals ▪ Convergence analysis is only focused on CDRs H1-H3 ▪ It is low-throughput and based on time intensive in vitro selections, due to Ab library being displayed on mammalian cells and sorted flow cytometry |