skDER and CiDDER: two scalable approaches for microbial genome dereplication
Bibliographic record
Abstract
ABSTRACT An abundance of microbial genomes have been sequenced in the past two decades. For fundamental comparative genomic investigations, where the goal is to determine the major gain and loss events shaping the pangenome of a species, it is often unnecessary and computationally onerous to include all available genomes in studies. In addition, over-representation of specific lineages due to sampling and sequencing bias can have undesired effects on evolutionary analyses. To assist users with genomic dereplication , selecting a subset of representative genomes, for downstream comparative genomic investigations, we developed skDER & CiDDER ( https://github.com/raufs/skDER ). skDER combines recent advances to efficiently estimate average nucleotide identity (ANI) between thousands of microbial genomes with two efficient algorithms for genomic dereplication. Further, CiDDER implements an approach whereby protein clusters are determined across all genomes and genomes are iteratively selected as representatives until a user-defined saturation of the total protein space is achieved. To support ease of use, several auxiliary functionalities are implemented within the two programs, including arguments to: (i) test the number of representative genomes resulting from a variety of clustering parameters, (ii) automate downloading of genomes belonging to a bacterial species or genus by name, (iii) cluster non-representative genomes to their closest representative genomes, and (iv) automatically filter predicted plasmids and phages prior to dereplication. We further assess the effects of filtering mobile genetic elements (MGEs) on ANI and alignment fraction (AF) estimates between pairs of genomes and find that MGEs tend to slightly deflate both metrics in one species. DATA SUMMARY skDER and CiDDER are provided as open-source software implemented in Python and C++ on Github: https://github.com/raufs/skDER ; with version updates tracked on Zenodo: https://zenodo.org/records/13887710 1 . Installation of the software is supported via both Bioconda 2 and Docker. Additional code and data for analyses presented in this manuscript can be found on Zenodo at: https://zenodo.org/records/13891800 3 . Pre-computed representative genomes selected by skDER (v1.0.7) for 18 common bacterial taxonomic groups, referencing classifications from GTDB release 214 4 , are also provided on Zenodo at: https://zenodo.org/records/10041203 5 . IMPACT STATEMENT Due to the increased availability of genomes for certain microbial species, performing fundamental comparative genomic investigations has become restricted to those with access to more advanced computational infrastructure. Genomic dereplication, the process of selecting distinct representative genomes to capture the breadth of a taxonomic group, thus presents a valuable solution to overcome this embarrassment of riches. Specifically, genomic dereplication allows simplifying the scale of comparative investigations while minimizing the risk of biasing analyses to specific lineages, which might be overrepresented in genomic databases. We present here two programs for genomic dereplication, one based on ANI-inference, skDER, and the other based on assessing the saturation of total coding-genes sampled, CIDDER. These tools are implemented with a variety of auxiliary options and designed for ease of use.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Direct model labels (unvalidated)
Per-model category and study-design labels from the labeling rounds. They are machine output, unvalidated, and the disagreement between models ships as data. No study design here is MEDLINE-validated yet.
| Model arm | Categories | Study design | Confidence |
|---|---|---|---|
| gemma | no category Domain: not available · Genre: Methods About the Canadian research system: no · About a Canadian topic: no | Simulation or modeling | low |
| gpt | no category Domain: not available · Genre: Methods About the Canadian research system: no · About a Canadian topic: no | Other design | high |
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.016 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.003 | 0.004 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.004 | 0.005 |
| Open science | 0.005 | 0.010 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.011 | 0.011 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedLabeled directly by 2 models reading the full record.
The models disagree on parts of this classification; every voice is preserved in the section at the end of the page.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".