Item request has been placed! ×
Item request cannot be made. ×
loading  Processing Request

Machine learning guided synthetic immunology and microbiology

Item request has been placed! ×
Item request cannot be made. ×
loading   Processing Request
  • معلومة اضافية
    • Contributors:
      Wong, Gerard C.L.
    • بيانات النشر:
      eScholarship, University of California
    • الموضوع:
      2024
    • Collection:
      University of California: eScholarship
    • نبذة مختصرة :
      This dissertation covers the implementation of computational algorithms to the area of microbiology and immunology. In chapter 1 and 2, we focus on the dynamic flagellar motilities performed by Pseudomonas aeruginosa. Using a combination of physical simulations and computer vision microscopy routines, we explore the diverse modes of individual and collective translation, not limited to free planktonic “swimming”. We show how bacteria can use their flagellar motor to explore surfaces and predict that flagellar instabilities may play an important role on surface detachment. Additionally, we interrogate the two stators, MotAB and MotCD, flagellar motor system of P. aeruginosa. We show how these homologous proton- driven stators set up flagellar intermittencies that affect the collective motility characteristic of P. aeruginosa. Beyond single cell motility propulsion, the stators may play a key role on intricate signaling network implemented by bacteria to sense its environment and decision making to transition from planktonic motility to biofilm formation. In chapter 3 and 4, we implement machine learning routines to identify antimicrobial peptide like, xenoAMP, motifs in enterotoxins, TcdA and TcdB, from Clostridioides difficile with immunomodulatory and membrane remodeling properties. Our results support a mechanism in which the large GTD participate on its own translocation in synergy with other domains including the DRBD. In addition to their membrane remodeling activity, we show that such toxin-derived peptides can organize dsDNA into pro-inflammatory complexes to trigger an amplified TLR9-meditated inflammatory response. Using a machine learning aided phylo-functional analysis, we find that TcdB variants from hypervirulent strains share similarities and high prediction scores at such xenoAMP motifs. Finally in chapter 5, we explore the world of milk proteins. We identify an immunomodulatory motif, AS2C30, in the αS2-casein protein fraction of cow and camel that is absent in human breast milk. The cow’s milk ...
    • Relation:
      qt35w177s2; https://escholarship.org/uc/item/35w177s2
    • Rights:
      public
    • الرقم المعرف:
      edsbas.D9945C7F