genome size

Deep research

Generated by just research-trait; source: research/traits/genomics/genome_size-deep-research-falcon.md

# Curation report: microbial genome size

**Target trait:** `traitmech:000098`  
**Label:** genome size  
**Category:** GENOMICS  
**Parent:** `METPO:1000188`  
**Status:** REVIEWED

## Executive summary

Microbial genome size is best curated as the **total length of the organism’s haploid genome**, in base pairs or megabase pairs. For bacteria and archaea, the record should state whether it includes only chromosomes or also plasmids, chromids, and other stable replicons. Genome size is an outcome of opposing processes: DNA acquisition, principally horizontal transfer and mobile-element integration, versus deletion, gene loss, and selection against costly or unnecessary functions.

The strongest general causal model is not a single “streamlining” pathway. At least two routes lead to small genomes: **drift-dominated reduction** following low effective population size, bottlenecks, weak recombination, and deletional bias; and **selection-dominated streamlining** in stable or nutrient-limited environments where accessory functions impose costs. Recent 2024 work challenges the assumption that enormous effective populations necessarily explain streamlined marine genomes, finding support for drift and mutation-rate effects in *Prochlorococcus* and Roseobacter lineages. These conclusions remain lineage-specific and partly model-based rather than universally established (wang2024aneutralprocess pages 1-5, zhang2024genomereductionoccurred pages 7-10, wang2024aneutralprocess pages 14-17, zhang2024genomereductionoccurred pages 10-14).

## 1. Trait scope and boundaries

### Operational definition

For TraitMech, `traitmech:000098` should represent:

> **The total number of nucleotide base pairs in one complete haploid complement of an organism’s genome, normally reported in bp, kbp, or Mbp.**

Recommended measurement fields are assembly accession, assembly status, estimated completeness/contamination, chromosome count, and whether plasmids/chromids are included. Complete isolates provide direct sequence length; metagenome-assembled genomes provide an **estimated genome size** that must be corrected or qualified for incompleteness.

### Distinguish from nearby traits

* **Gene count or coding capacity:** closely correlated with bacterial genome size but not identical. Pseudogenes, intergenic DNA, repeats, and MGEs can change length without proportionate changes in functional genes.
* **Pangenome size:** the union of genes across strains of a species, not the length of an individual genome.
* **Assembly span:** incomplete MAGs and fragmented draft assemblies systematically underestimate true genome size.
* **DNA content and ploidy:** flow-cytometric DNA per cell varies with chromosome copy number and replication state; it is not necessarily haploid genome length.
* **Chromosome size:** excludes plasmids and secondary replicons unless explicitly included.
* **Minimal genome:** an experimentally defined set sufficient under specified conditions, not the minimum naturally possible genome size. JCVI-syn3.0, for example, has 531,560 bp and 473 genes, but its viability depends on a rich laboratory environment (hutchison2016designandsynthesis pages 5-6).
* **Cell size:** potentially correlated in some taxa, but it is a distinct morphology trait and should not be treated as a proxy.

### Boundary cases

1. **Multipartite genomes:** record chromosome-only and total-replicon lengths separately where possible.
2. **Integrated prophages and genomic islands:** count them when integrated into the sequenced chromosome; retain MGE annotations as explanatory nodes.
3. **Transient plasmids:** inclusion can make genome size condition- or strain-dependent. Curate replicon policy explicitly.
4. **Polyploid archaea/bacteria:** report haploid sequence length, not total cellular DNA.
5. **Endosymbionts and uncultivated taxa:** strong incompleteness and contamination controls are essential because the most reduced genomes are particularly vulnerable to assembly artifacts.

## 2. Current mechanistic understanding

### Drift-dominated reduction

Low effective population size weakens purifying selection. Slightly deleterious gene inactivation, pseudogenization, and mobile-element expansion can therefore persist; bacterial deletional bias subsequently erodes nonfunctional DNA. Low recombination can compound this process through Muller’s ratchet. In early *Prochlorococcus*, modeling placed the relevant effective-population-size range near **10⁴–10⁵**, comparable to obligate endosymbionts, and inferred drift as the principal historical driver. The same study reported *Prochlorococcus* recombination-to-mutation ratios of approximately **1–3**, versus **61–63** for SAR11, consistent with weaker removal of deleterious variation (zhang2024genomereductionoccurred pages 7-10, zhang2024genomereductionoccurred pages 10-14).

A 2024 Roseobacter preprint used long-term mutation-accumulation experiments and **437 mutant lines** spanning 2–3, 3–4, and 4–5 Mb genome groups. Effective population size scaled positively with genome size and mutation rate scaled negatively, contrary to the simplest streamlining-selection prediction. The authors interpret drift as the ultimate driver, although the evolutionary conclusion combines experiments with population-genetic inference and remains preprint evidence (wang2024aneutralprocess pages 1-5, wang2024aneutralprocess pages 14-17).

### Selection-dominated streamlining

In stable or specialized environments, genes useful only in alternative niches can become costly. Selection may favor deletion of these accessory functions rather than favoring shorter DNA per se. In *Methylobacterium extorquens*, 1,500 generations of experimental evolution produced nearly parallel deletions in **80% of populations**, removing up to **10% of a megaplasmid**. Reconstructed deletions were beneficial in the selected environments but impaired performance elsewhere; reported fitness gains were **14.5–26.0%** (lee2012repeatedselectiondrivengenome pages 7-8).

Nutrient limitation can also favor low biosynthetic demand and efficient resource allocation. However, recent work indicates that present-day benefits of a small genome do not prove that selection originally caused the reduction. In *Prochlorococcus*, small genomes may now improve nutrient and light assimilation while retaining signatures of drift-dominated history (zhang2024genomereductionoccurred pages 10-14).

### DNA acquisition and genome expansion

Horizontal gene transfer by plasmids, integrative elements, transposons, insertion sequences, and bacteriophages adds accessory DNA and can expand metabolic range, resistance, virulence, or environmental tolerance. Net genome-size effects are conditional: acquired elements can be maintained under selection, become pseudogenized, or subsequently be deleted. Consequently, “HGT increases genome size” is suitable only as an event-level edge—**integration/acquisition adds DNA**—not as a universal long-term ecological rule.

Showing the first 60 of 269 lines of findings; the linked file also carries the run's front matter and the prompt it was given — read the full report.