That is the challenge researchers face when studying how bacteria adapt to different environments. A bacterial gene may appear to be linked to a particular habitat, when the real reason is simply that closely related bacteria happen to live there.

In their 2025 Genome Biology paper, Bujdoš, Walter, and O’Toole introduce aurora, a machine-learning tool designed to tackle this problem.

Aurora identifies potentially mislabeled or unusual bacterial strains before performing genome-wide association studies (GWAS). By cleaning up the dataset and accounting for bacterial evolutionary relationships, it can help researchers find genetic features that are more genuinely connected to habitat adaptation.

The researchers tested aurora using simulated and real bacterial datasets and found that it could recover important genetic associations even when datasets contained misleading labels.

The bigger lesson is simple: better biological discoveries often begin with better data.

Aurora gives researchers a new way to separate real genetic clues from misleading ones—and could help us better understand how bacteria adapt, survive, and evolve in the environments they call home.

Based on Bujdoš et al., “aurora: a machine learning GWAS tool for analyzing microbial habitat adaptation,” Genome Biology (2025).[Read the original paper](https://link.springer.com/article/10.1186/s13059-025-03524-7?)

Research Directions

Stony corals: evolution of novelty and photosymbiosis

Symbiotic acoels: cell type evolution and photosymbiosis

Animal genomes: structural evolution and gene regulation

https://sgel.biodiv.tw/home

A recent study published in Genome Biology (https://link.springer.com/article/10.1186/s13059-026-04280-y) addressed this challenge by introducing dicast (https://github.com/burgshrimps/dicast), a machine-learning approach for detecting structural variants in short-read sequencing data. Structural variants, such as large deletions, insertions, duplications, and inversions, can have significant biological and clinical implications, yet they are challenging to identify with short-read technologies. Because different detection methods frequently yield divergent results, researchers commonly employ consensus calling, considering a variant valid only if multiple tools detect it. While this approach reduces false positives, it relies on the potentially flawed assumption that the majority is always correct.

The researchers explored the impact of evaluating the supporting evidence for each variant, rather than simply tallying the number of algorithms that identified it. To establish a ground truth, they analyzed nine genomes using multiple sequencing technologies and 15 detection methods, initially identifying approximately 35 million potential variants. Through extensive filtering, evidence integration, and manual review of over 11,500 variants, they developed a robust benchmark comprising more than 236,000 structural variants. The findings underscored the complexity of the problem: short-read methods detected fewer than half of deletions and less than 10 percent of insertions, with performance declining markedly in repetitive genomic regions. In contrast, long-read technologies demonstrated superior detection capabilities. However, replacing the substantial volume of existing short-read data in clinical and research settings is not immediately feasible. Consequently, the researchers questioned whether short-read data might harbor more information than conventional analytical pipelines currently extract.

This line of inquiry led to the development of dicast. Rather than merely confirming agreement among multiple tools, dicast identifies patterns in sequencing data, including split and clipped reads, discordant read pairs, alignment characteristics, and the surrounding genomic context. An XGBoost machine-learning model evaluates which combinations of these signals are indicative of genuine structural variants. Thus, the approach shifts from tallying algorithmic consensus to interpreting the underlying evidence.

The researchers subsequently conducted a targeted evaluation by examining structural variants detected by only a single short-read tool, which are typically missed by consensus-based approaches. dicast successfully recovered approximately 81% of these single-caller deletions, insertions, and duplications. The signals for these variants were present in the data, but conventional filtering methods failed to integrate them effectively.

The utility of dicast was further demonstrated in cohorts with rare diseases, including congenital limb malformations, atrial fibrillation, and neuromuscular disorders. In one instance, dicast achieved a deletion recall rate of approximately 0.96, compared to 0.74 using consensus calling. The median number of variants requiring manual review was 29 per sample. Among 31 experimentally validated variants that standard filters would have missed, dicast identified 12, whereas consensus calling detected only one.

Overall, dicast identified approximately 20 percent more potential disease-causing deletions than consensus-based methods. While a 20 percent increase may appear modest, in clinical genomics such improvements can have significant practical implications. Missing a deletion may leave a case unresolved, whereas detecting a structural variant can provide critical diagnostic insights.

The study does not claim that machine learning has rendered short-read sequencing superior to long-read approaches. Instead, the results underscore the effectiveness of long-read sequencing for structural variant detection. However, dicast highlights a more nuanced perspective: substantial biological information may still be recoverable from the extensive short-read datasets already available.

The principal lesson extends beyond the detection of structural variants. For many years, bioinformatics pipelines have relied on threshold-based criteria, such as minimum coverage, quality scores, or support from multiple tools. While these rules are useful, biological phenomena do not always conform to rigid checklists; multiple weak signals, when considered collectively, can provide compelling evidence.

This perspective prompts consideration of the solitary variant: one algorithm identifies it, while several others do not. Traditional consensus techniques might have dismissed it, yet machine learning approaches evaluate the available evidence to determine whether the variant is plausible.

Occasionally, the most significant variant within a genome is the one that is almost universally overlooked.

Read more at https://link.springer.com/article/10.1186/s13059-026-04280-y

Address of the bookmark: https://github.com/chhylp123/hifiasm

Address of the bookmark: http://genome.cs.nthu.edu.tw/Multi-CAR/

So far miniasm is in early development stage. It has only been tested on a dozen of PacBio and Oxford Nanopore (ONT) bacterial data sets. Including the mapping step, it takes about 3 minutes to assemble a bacterial genome. Under the default setting, miniasm assembles 9 out of 12 PacBio datasets and 3 out of 4 ONT datasets into a single contig. The 12 PacBio data sets are PacBio E. coli sample, ERS473430, ERS544009, ERS554120, ERS605484, ERS617393, ERS646601, ERS659581, ERS670327, ERS685285, ERS743109 and a deprecated PacBio E. coli data set. ONT data are acquired from the Loman Lab.

For a C. elegans PacBio data set (only 40X are used, not the whole dataset), miniasm finishes the assembly, including reads overlapping, in ~10 minutes with 16 CPUs. The total assembly size is 105Mb; the N50 is 1.94Mb. In comparison, the HGAP3produces a 104Mb assembly with N50 1.61Mb. This dotter plot gives a global view of the miniasm assembly (on the X axis) and the HGAP3 assembly (on Y). They are broadly comparable. Of course, the HGAP3 consensus sequences are much more accurate. In addition, on the whole data set (assembled in ~30 min), the miniasm N50 is reduced to 1.79Mb. Miniasm still needs improvements.

Miniasm confirms that at least for high-coverage bacterial genomes, it is possible to generate long contigs from raw PacBio or ONT reads without error correction. It also shows that minimap can be used as a read overlapper, even though it is probably not as sensitive as the more sophisticated overlapers such as MHAP and DALIGNER. Coupled with long-read error correctors and consensus tools, miniasm may also be useful to produce high-quality assemblies.

Minimap and miniasm are ultrafast tools for (i) mapping and (ii) assembly. Designed for long, noisy reads, they do not have a correction or consensus step, and therefore the resulting assemblies are contiguous (i.e. long) but very noisy (i.e. full of errors)

We start with an all against all comparison:

minimap -Sw5 -L100 -m0 -t8 reads.fq reads.fq | gzip -1 > reads.paf.gz

Then we can assemble

miniasm -f reads.fq reads.paf.gz > reads.gfa

Convert GFA to FASTA:

awk '/^S/{print ">"$2"\n"$3}' reads.gfa | fold > reads.fa

And then count how many contigs:

grep ">" reads.fa | wc -l

# Download sample PacBio from the PBcR website
wget -O- http://www.cbcb.umd.edu/software/PBcR/data/selfSampleData.tar.gz | tar zxf -
ln -s selfSampleData/pacbio_filtered.fastq reads.fq
# Install minimap and miniasm (requiring gcc and zlib)
git clone https://github.com/lh3/minimap && (cd minimap && make)
git clone https://github.com/lh3/miniasm && (cd miniasm && make)
# Overlap
minimap/minimap -Sw5 -L100 -m0 -t8 reads.fq reads.fq | gzip -1 > reads.paf.gz
# Layout
miniasm/miniasm -f reads.fq reads.paf.gz > reads.gfa

Address of the bookmark: https://github.com/lh3/miniasm

Please see Achieving Success with De Novo Assembly and System Requirements before creating your Chromium libraries for assembly.

Supernova should be run using 38-56x coverage of the genome.
• Somewhat higher coverage is sometimes advantageous.
• Supernova will exit if it finds that coverage is far from the recommended range.
• Note that at most 2.14 billion reads are allowed.
• Please note that we have not extensively tested genomes larger than human, and any genome above approximately 4 GB should be considered experimental and is not supported.

Address of the bookmark: https://support.10xgenomics.com/de-novo-assembly/software/pipelines/latest/using/running

install.packages("gggenes")

If you want the development version, install it from GitHub:

devtools::install_github("wilkox/gggenes")

More at https://github.com/wilkox/gggenes

Address of the bookmark: http://wilkox.org/gggenes