UC Santa Barbara

Horizontal Gene Transfer (HGT) refers to sharing of small segments of genetic material from a donor to a recipient organism that does not have a parent-offspring relationship. Although not all transfers are successful, they are still abundant in natural bacterial communities. Moreover, the rate at which a gene is transferred differs widelyacross different genes and organisms. This makes it challenging, especially for healthcare professionals that study antimicrobial resistance and epidemics, to predict how the organism will evolve over time and over environmental fluctuations. Recombination creates more opportunities for a local adaptation by making it possible to acquire genes involved in antibiotic resistance, pathogenic determinants, etc. Many metagenomic studies have used natural biofilms in order to measure the rates of gene transfer across communities. Moreover, not only can one study genes responsible for protection against outside influence or metabolism but metagenomic studies of biofilms and microbial mats can provide invaluable insights into evolutionary processes within natural communities. In the first chapter a comprehensive analysis of the Pink Berry metagenomic data is performed in context of extensive recombination of a quasi-sexual bacterial population. Evidence is presented showing that the populations are divided into clades with different evolutionary histories including a mixing layer where bacteria experience extensive recombination from other clades as well as with each other. In the second part of this thesis I focus on self / non-self bacterial recognition. Toxinantitoxin systems are important mechanisms for the bacteria to respond to intracellular stress. Usually, the proteins that are a part of the contact dependent growth inhibition (CDI) secretion contain multiple distinct parts - the structure that acts as a delivery mechanism, the toxin, and the antitoxin. Here the preliminary findings of diversity in WapA and RhsC C-termini in purple sulfur bacteria is shown. The presence of large gaps in the alignment of the C-terminus region of the CDI proteins shows that the bacteria differ in their repertoire of toxins depending on their geographical location. These observations mark a promising starting point for studying CDI mechanisms in naturally occurring bacterial populations. Lastly, the composition of bacterial communities grown in different conditions based on their 16S sequences is analyzed. In this chapter I have introduced a new tool for decreasing the error of 16S Nanopore sequences and identifying given samples. The accuracy of the pipeline using simulated PacBio and Nanopore datasets from CAMI2 is demonstrated and then it is used on experimental sediment data.