Page 130 - Read Online

Wang et al. Microbiome Res Rep. 2025;4:23 https://dx.doi.org/10.20517/mrr.2024.94 Page 15 of 19

Among the various spike-in absolute quantification methods, using extracted DNA as a spike-in instead of
directly adding cells to fecal samples offers significant advantages. This approach eliminates biases
associated with cell lysis efficiency and DNA extraction recovery [49,50] , ensuring more reliable quantification.
Additionally, for samples with pre-extracted DNA, quantification can begin directly from the DNA,
bypassing the need for additional sample preparation steps. Compared to spike-in synthetic DNA
sequences, which may introduce artificial background noise or biases in bioinformatic analyses during
sequencing , using authentic biological DNA, such as marine bacterial DNA, better mimics human gut
[18]
microbiome DNA. This reduces potential interference caused by sequence discrepancies and enhances the
accuracy of microbial community profiling.

The use of copy number as a standard can result in amplification efficiency errors during 16S rRNA
[51]
sequencing, affecting total copy number estimation . This issue is particularly pronounced in samples with
a high abundance of high-copy-number bacteria. However, this impact should be minimized in shotgun
in absolute abundance analysis (F = 6.3) suggests that the relative abundance analysis exhibits a slightly
sequencing, as it involves shearing DNA rather than amplification [51,52] .

The observed shifts in genus rankings between relative and absolute quantification [Figure 4C-E] are driven
by both total microbial load differences and individual taxon variation, with total load playing the primary
role. This effect is particularly evident between infant and mother samples, where mothers had significantly
higher total bacterial counts, amplifying the contribution of dominant maternal genera in absolute terms.
For instance, while Bifidobacterium appeared dominant in infants based on relative abundance, its absolute
copy number was comparable between infants and mothers, indicating a load-driven distortion. In contrast,
Escherichia-Shigella remained significantly more abundant in infants even after normalization, reflecting
true taxon-specific enrichment. These results underscore the importance of absolute quantification in
disentangling microbial abundance from compositional bias. Without adjusting for total load, key ecological
signals may be obscured or misrepresented in relative abundance–based analysis.

When comparing relative abundance with absolute abundance after spike-in processing, it was found that
Veillonella, Streptococcus, and Enterobacter are no longer among the top 10 in the absolute quantification.
Veillonella's biofilm-forming ability , Streptococcus’s transformation capability , and Enterobacter's drug
[53]
[54]
[55]
efflux mechanisms confer antibiotic resistance to these genera. Consequently, when participants take
antibiotics, the total number of bacterial cells decreases , and antibiotic-resistant genera become more
[56]
[16]
prominent in relative quantification analyses. This finding is consistent with previous studies .
The F value in both analyses indicates the ratio of the variance between the groups to the variance within
the groups . The slightly higher F value in the relative abundance analysis (F = 6.6) compared to the spike-
[57]
greater distinction between the mother and infant microbiomes. This difference in F values may be
attributed to the inherent variability and potential biases in relative abundance measures, which can
sometimes amplify differences between groups. In contrast, the spike-in method aims to reduce such biases
by providing an absolute quantification, potentially leading to a more accurate but slightly lower F value.

This interpretation is further supported by taxonomic patterns observed in Figure 4, where high-biomass
maternal taxa such as Blautia and Coprococcus and infant-associated taxa like Escherichia-Shigella and
Bifidobacterium largely drove the separation in both relative and absolute β-diversity plots. Genera like
Veillonella, although prominent in relative terms, were low in absolute abundance and contributed less to
the spike-in–based analysis. The similar PCoA clustering patterns and consistent variance explained across
PC1 and PC2 further support that absolute β-diversity reflects the same underlying biological differences -