The distinction between dry and wet preservation carries consequences that extend well beyond the curatorial preferences of any given era. When organisms are fixed in formalin or similar crosslinking agents, the chemistry that hardens soft tissues for anatomical study also attacks the molecule that genomic researchers care about most. Formaldehyde reacts with amino groups in DNA and associated proteins, forming methylene bridges that lock nucleic acids into an insoluble tangle. Even when specimens are transferred from formalin into ethanol for long-term storage, the initial fixation period can be enough to inflict damage that no later treatment can undo. Ethanol itself, while gentler, is not harmless: dehydration removes water molecules that normally cushion the DNA double helix, and over decades the combined stresses of desiccation, oxidation, and hydrolysis steadily shear the genome into ever shorter pieces. The result is that a fluid-preserved specimen from the early twentieth century may hold DNA fragments averaging well under a hundred base pairs, a stark contrast to the tens of thousands of base pairs routinely recovered from fresh tissue.

This fragmentation is why the choice of extraction protocol matters so much for wet specimens. Standard silica-column kits, which were engineered with fresh blood or tissue in mind, tend to lose much of the already scarce material from historical samples. Short fragments bind inefficiently to the silica membrane under standard conditions, and aggressive wash steps can wash away what little has bound. Formalin-focused kits attempt to reverse some of the crosslinking chemistry with modified lysis buffers and extended digestion, while bespoke homebrew protocols give researchers the freedom to tune incubation times, detergent concentrations, and binding conditions to the peculiar chemistry of a given specimen. The comparison of commercial and custom approaches in this study therefore speaks to a practical question that collections managers and molecular ecologists face constantly: which route recovers the most usable genetic material from a finite and irreplaceable tissue sample?

The study system itself deserves attention. Crayfish are an excellent test case for museomic methods because their exoskeletons and gills present particular challenges and opportunities. Gills, the tissue chosen here based on prior recommendations for crayfish genomics, combine soft tissue accessibility with relatively high cell density, and they can often be sampled with minimal disturbance to external morphology that taxonomists rely on. Decapod crustaceans also have notoriously large and complex genomes, which amplifies the difficulty of recovering informative sequence from degraded material. If an extraction method performs well on crayfish gill tissue, it is likely to be a reasonable candidate for other fluid-preserved invertebrates with similarly challenging genomes. The rusty crayfish, an invasive species in much of the Midwestern United States with a well-documented expansion history, also illustrates the kind of population-level questions that museum genomics can address: historical specimens collected before or during the early stages of invasion provide a genetic baseline that no amount of contemporary sampling can recreate.

The finding that sample age reduces DNA yield across all extraction methods is consistent with a large body of museomic literature, but the consistency of the pattern across protocols is itself informative. It suggests that the dominant driver of degradation is the cumulative chemical history of the specimen rather than any particular quirk of a given extraction chemistry. Time in fixative, exposure to fluctuating storage conditions, and the slow hydrolysis of phosphodiester bonds all proceed regardless of which kit is used to liberate the DNA afterward. This has a sobering implication for collection managers: the best time to preserve genomic information is at the moment of accession, and every year of suboptimal storage narrows the window of what future researchers can recover. It also means that the oldest specimens, often the most scientifically valuable because they document populations and environments that no longer exist, are precisely the ones from which the least DNA can be pulled.

The finding that starting tissue mass had no significant effect on DNA yield is more surprising and more encouraging. Intuitively, one might expect that doubling the amount of gill tissue would double the recovered DNA, but the results suggest that once a minimum threshold of material is present, additional mass does not translate into additional recoverable molecules. This may reflect saturation of the extraction chemistry, or it may reflect the reality that in heavily degraded samples the limiting factor is the number of intact DNA molecules per unit tissue rather than the total amount of tissue offered. Either way, the practical consequence is favorable for curators: if a small snip of gill tissue performs as well as a larger one, then the destructive sampling burden on each specimen can be minimized without sacrificing genomic output. This aligns with the growing ethic of minimal-impact sampling in natural history collections, where the goal is to extract maximum information while preserving the specimen for future techniques that may require even less material.

The variability in loci recovery across extraction methods, even when base call accuracy remained high, highlights a subtlety of reduced-representation sequencing that deserves elaboration. Genotyping-by-sequencing depends on a restriction enzyme cutting the genome at specific recognition sites. If the DNA fragments in a sample are shorter than the distance between adjacent restriction sites, many fragments will lack a cut site entirely and cannot be captured by the protocol. As average fragment length shrinks with specimen age and preservation damage, the effective number of loci that can be recovered falls in a nonlinear fashion. Two extraction methods that produce similar total DNA quantities may nevertheless produce very different fragment length distributions, and this difference is invisible in a simple yield measurement but immediately apparent in the number of loci recovered after sequencing. This is why the study’s decision to evaluate sequencing quality alongside raw yield is methodologically important: yield alone is an incomplete proxy for the utility of an extraction for downstream genomic applications.

Base call accuracy, by contrast, appears more resilient. Even when loci recovery was variable, the sequences that were recovered could be read with confidence, suggesting that the chemical damage present in these specimens does not systematically mislead the sequencer. This is good news for researchers working with fluid-preserved material, because it means that the data produced from wet specimens can be trusted once sufficient loci are obtained. It also suggests that the primary bottleneck is quantity and length of recoverable fragments rather than the fidelity of the bases themselves, which points optimization efforts toward extraction chemistry rather than sequencing chemistry.

The broader context of museomics makes this kind of methodological work increasingly urgent. Natural history collections worldwide hold billions of specimens, a substantial fraction of which are fluid-preserved, and the genomic information locked inside them represents a temporal depth that no contemporary sampling program can match. As genomic datasets grow in importance for conservation planning, phylogenetics, and the study of rapid environmental change, the ability to tap this archive becomes more valuable. At the same time, specimens are finite, and destructive sampling decisions must be weighed carefully. Studies like this one, which systematically compare extraction approaches on a well-controlled set of specimens spanning more than a century, provide the empirical foundation that curators need to make informed decisions about which specimens to sample and which methods to authorize.

The historical contingency of preservation practices adds another layer of complexity. Because fixative recipes, fixation durations, and transfer protocols varied widely across institutions and decades, two specimens of the same age in different collections may have very different genomic prospects. A specimen fixed briefly in buffered formalin and promptly transferred to ethanol may retain far more usable DNA than one left in unbuffered formalin for years, even if both entered the collection the same year. The acidic hydrolysis that unbuffered formalin promotes is particularly damaging, and many older collections contain specimens whose fixation history is poorly documented or entirely unknown. This uncertainty means that researchers often cannot predict in advance how well a given specimen will perform, and pilot extractions or non-destructive sampling approaches may be warranted before committing rare material to a full protocol.

The comparison of commercial kits against a custom homebrew protocol also reflects a broader tension in the field. Commercial kits offer reproducibility, quality control, and ease of use, which matters when many specimens must be processed consistently. Custom protocols offer flexibility and the potential for optimization to particular preservation histories or tissue types, but they introduce variability between laboratories and require more expertise to implement correctly. The finding that different methods produce different loci recovery profiles suggests that neither approach is universally superior, and that the best choice depends on the specific specimens, the genomic questions being asked, and the resources available. For laboratories working extensively with fluid-preserved material, the takeaway may be that a hybrid approach, using commercial kits as a baseline and custom modifications where needed, offers the best balance of reliability and performance.

Looking forward, the continued development of extraction methods tailored to degraded and chemically modified DNA will only increase the value of fluid-preserved collections. Techniques borrowed from ancient DNA research, including single-stranded library preparation and uracil-tolerant polymerases, are already being adapted for museum genomics, and further refinements may unlock material that is currently considered too degraded to be useful. Every improvement in extraction efficiency effectively extends the temporal reach of existing collections, allowing specimens collected in the nineteenth century to contribute to questions posed in the twenty-first. In this sense, methodological studies of DNA capture from wet specimens are not merely technical exercises but investments in the future utility of the biodiversity archive that natural history museums have been quietly accumulating for centuries.

Subject of Research: …

Article Title: We have specimens at home: efficacy comparison of methods for genomic DNA capture in fluid-preserved crayfish specimens

Article References: Wallace, J. R., Skelly, Q. M., Carstens, B. C., & Daly, M. (2026). We have specimens at home: efficacy comparison of methods for genomic DNA capture in fluid-preserved crayfish specimens. Natural History Collections and Museomics, 3, Article e189111. https://doi.org/10.3897/nhcm.3.189111

Image Credits: AI Generated

DOI: 10.3897/nhcm.3.189111

Keywords: specimens, home, efficacy, comparison, methods, genomic, capture, fluid-preserved, crayfish, scientific research