UC Berkeley

The core machinery required for oxygenic photosynthesis is well conserved from cyanobacteria to land plants and in various algal clades, and has been studied intensely for decades. Much remains to be learned, however, both about the regulation and assembly of the core machinery, and about a host of optional factors associated with it only in particular lineages or under particular conditions. I sought to identify unstudied genes required for long-term acclimation to high light intensity in the cyanobacterium Synechococcus elongatus PCC 7942 by simultaneously comparing the growth rates of mutants in every nonessential gene in the genome under low, normal, and high light via RB-TnSeq (growth of a pooled transposon library, followed by sequencing and quantification of DNA barcodes). This was the first study to investigate low and high light conditions with high enough resolution to as- sign fitness values to nearly every gene. It confirmed that high light mainly damages the Photosystem II reaction center, and that synthesis and recycling of chlorophyll from damaged PSII subunits is essential for acclimation. It also identified several unexpected genes with large fitness effects. The include a mys- terious intrinsically disordered protein with homology to phasins, an unusual transcriptional regulator that may physically protect DNA from oxidative damage, and an ABC transporter that takes up amino acids. Next I grew the transposon library under a range of fluctuating light conditions to determine which genes are important for fitness only when fluctuations include higher highs or lower lows, alternate quickly or slowly, include gradual or sudden low-to-high transitions, or mimic the patterns found in dense bioreactor cultures. Many genes were beneficial only under certain conditions but not others, and some were also found to be detrimental (knocking them out increased fitness) under certain con- ditions. The genes identified are implicated in diverse cellular processes relevant to photosynthesis, in- cluding: the stringent response to dark periods, alternative electron sinks (plastoquinone reduction, glycogen synthesis) needed during sudden shifts to high light, and regulation of transcription. As part of a separate line of inquiry into genes conserved across photosynthetic species but missing from other branches of life, I also developed a scripting language that partially automates comparisons between large numbers of genomes using a variety of popular sequence search and ortholog-finding programs: BLAST, BLAST reciprocal best hits, HMMER, DIAMOND, MMSeqs2, OrthoFinder, SonicParanoid, and others. It reproducibly installs all the programs and runs them as needed to per- form searches as described in a concise, human-readable domain-specific language designed to facilitate sharing and incremental improvement of the search algorithms.