Horizontal gene transfer (HGT), acting at various phylogenetic scales, is responsible for major evolutionary innovations – particularly in microbes. Detecting HGT, particularly at local/population genomic scales, is a difficult and potentially computationally – and statistically – challenging problem. We have yet to settle on a set of comprehensive and generalizable methods for detecting recombination and HGT in bacterial genomes so that these patterns can be combined with signatures of natural selection to address evolutionary responses in ecologically relevant traits in natural bacterial populations. In Rhizobium symbionts of legumes, for example, the region of highest differentiation between control and N-evolved strains (which have become less beneficial for plant hosts) appears to have been transmitted across a diverse set of chromosomal lineages. Also, another legume symbiont, Sinorhizobium meliloti, exhibits an intricate pattern of gene transference among chromosomal lineages and replicons (chromosome and megaplasmids), which involves homologous recombination and gene loss. Addressing these problems with rigor requires integrative work combining multiple classical and novel methods. Using the genomes of hundreds of rhizobial strains, GEMS postdoc Mario Cerón Romero is working with Katy Heath and Tandy Warnow to develop novel pipelines to detect recombination using both long read and short read genomes. This approach will be applied to Bombella and Pseudomonas, and beyond, as part of ongoing research in the GEMS Institute.