# Comparison of Spatio‐Temporal Dynamics and Composition in Size‐Fractionated and Unfractionated Northwestern Atlantic Microbial Communities

**Authors:** Diana Haider, Jennifer Tolman, Robert G. Beiko, Julie LaRoche

PMC · DOI: 10.1111/1758-2229.70206 · Environmental Microbiology Reports · 2026-02-06

## TL;DR

This study compares microbial communities in size-fractionated and unfractionated ocean samples, showing that pooling fractionated samples can reproduce unfractionated profiles, aiding cross-study comparisons.

## Contribution

The study demonstrates that pooling size-fractionated marine samples before or after sequencing yields community profiles nearly identical to unfractionated samples.

## Key findings

- De-fractionated samples closely resemble unfractionated samples in composition and richness, except for very rare taxa.
- Size fractionation impacts community composition more during phytoplankton blooms than before.
- Differential-abundance analysis found minimal differences between fractionated and de-fractionated samples.

## Abstract

Size fractionation, filtering water sequentially through a large (3 μm) and fine (0.2 μm) pore size filter, is a widely applied approach to target specific microbial size ranges, differentiating between particle‐associated, and free‐living microorganisms. To characterise its impact on microbial diversity and its comparability to unfractionated samples, we analysed 16 weekly ocean samples across five depths during a phytoplankton spring bloom. We used a universal marker to characterise prokaryotes, eukaryotes and chloroplasts in unfractionated (> 0.2 μm), fractionated (large > 3 μm, small 0.2–3 μm) and de‐fractionated samples, the reconstitution of the small and large fractions. The particle‐associated fraction was the most different community from all other samples, and de‐fractionating before or after sequencing results in a community that is most similar to unfractionated samples in terms of composition and richness with the exception of very rare taxa. Across all depths and weeks, 75%–97% of ASVs were shared, but some discrepancies in relative abundances were unresolved, including for some lineages of free‐living Bacteroidota. Community composition differences from size fractionation were more pronounced during the bloom period in comparison to pre‐bloom. Differential‐abundance analysis detected at most one significantly different ASV between fractionated and de‐fractionated samples, highlighting the similarity in community composition and temporal dynamics between the fractionated and de‐fractionated sets.

Size fractionation differentiates large, particle‐associated microbes from small, free‐living ones. We show that pooling size‐fractionated marine samples, normalised by their DNA concentration, either before or after sequencing produces community profiles nearly identical to unfractionated samples. This provides a strong basis for comparing samples obtained from different filtration methods or different research projects. Size fractionation reveals niche‐specific associations that might otherwise be overlooked, enhancing our ability to detect interactions between small and large microorganisms during dynamic periods like spring blooms.

## Full-text entities

- **Diseases:** W (MESH:C538106), SL (MESH:D008224)
- **Chemicals:** oxygen (MESH:D010100), carbon (MESH:D002244), phosphate (MESH:D010710), water (MESH:D014867), L (MESH:D007930), nitrogen (MESH:D009584), W (MESH:D014414), ammonia (MESH:D000641), S (MESH:D013455), nitrate (MESH:D009566), silicate (MESH:D017640), Buffer AP1 (-)
- **Species:** Moritella (genus) [taxon 58050], Bacteroides (genus) [taxon 816], Chaetoceros (genus) [taxon 49237], Parvimonas (genus) [taxon 543311], Fluviicola (genus) [taxon 332102]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12881702/full.md

## Figures

6 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12881702/full.md

## References

55 references — full list in the complete paper: https://tomesphere.com/paper/PMC12881702/full.md

---
Source: https://tomesphere.com/paper/PMC12881702