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

In the mid 1960's scientists discovered that many genomes contain stretches of highly repetitive DNA sequences ( see Reassociation Kinetics Experiments, and C-Value Paradox ). These sequences were later characterized and placed into five categories:

Simple Repeats - Duplications of simple sets of DNA bases (typically 1-5bp) such as A, CA, CGG etc.
Tandem Repeats - Typically found at the centromeres and telomeres of chromosomes these are duplications of more complex 100-200 base sequences.
Segmental Duplications - Large blocks of 10-300 kilobases which are that have been copied to another region of the genome.
Interspersed Repeats
Processed Pseudogenes, Retrotranscripts, SINES - Non-functional copies of RNA genes which have been reintegrated into the genome with the assitance of a reverse transcriptase.
DNA Transposons
Retrovirus Retrotransposons
Non-Retrovirus Retrotransposons ( LINES )

Currently up to 50% of the human genome is repetitive in nature and as improvements are made in detection methods this number is expected to increase.

On the other hand; In genetics, the term palindrome refers to a sequence of nucleotides along a DNA (deoxyribonucleic acid) or RNA (ribonucleic acid) strand that contains the same series of nitrogenous bases regardless from which direction the strand is analyzed. Akin to a language palindrome—wherein a word or phrase is spelled the same left-to-right as right-to-left (e.g., the word RADAR or the phrase "able was I ere I saw elba")—with genetic palindromes it does not matter whether the nucleic acid strand is read starting from the 3' (three prime) end or the 5' (five prime) end of the strand.

Recent research on palindromes centers on understanding palindrome formation during gene amplification. Other studies have attempted to relate palindrome formation to molecular mechanisms involved in double stranded breaks and in the formation of inverted repeats. Assisted by high speed computers, other groups of scientists link palindrome formation to the conservation of genetic information.

Related to the direction of transcription by RNA polymerase, DNA strands have upstream and downstream terminus defined by differing chemical groups at each end. The ends of each strand of DNA or RNA are termed the 5' (phosphate bound to the 5' position carbon) and 3' (phosphate bound to the 3' carbon) ends to indicate a polarity within the molecule. Using the letters A, T, C, G, to represent the nitrogenous bases adenine, thymine, cytosine, and guanine found in DNA, and the letters A, U, C, G to represent the nitrogenous bases adenine, uracil, cytosine, guanine found in RNA (Note that uracil in RNA replaces the thymine found in DNA), geneticists usually represent DNA by a series of base codes (e.g., 5' AATCGGATTGCA 3'). The base codes are usually arranged from the 5' end to the 3' end.

Because of specific base pairing in DNA (i.e., adenine (A) always bonds with (thymine (T) and cytosine (C) always bonds with guanine (G)) the complimentary stand to the sequence 5' AATCGGATTGCA 3' would be 3' TTAGCCTAACGT 5'.

With palindromes the sequences on the complimentary strands read the same in either direction. For example, a sequence of 5' GAATTC3' on one strand would be complimented by a 3' CTTAAG 5' strand. In either case, when either strand is read from the 5' prime end the sequence is GAATTC. Another example of a palindrome would be the sequence 5' CGAAGC 3' that, when reversed, still reads CGAAGC.

Palindromes are important sequences within nucleic acids. Often they are the site of binding for specific enzymes (e.g., restriction endobucleases) designed to cut the DNA strands at specific locations (i.e., at palindromes).

Palindromes may arise from brakeage and chromosomal inversions that form inverted repeats that compliment each other. When a palindrome results from an inversion, it is often referred to as an inverted repeat. For example, the sequence 5' CGAAGC 3', if inverted (reversed 180°), still reads CGAAGC.

The European Molecular Biology Open Software Suite (EMBOSS) includes some basic tools for finding tandem repeats and inverted repeats (see B.6.22. Applications in group Nucleic:repeats). There are many on-line services providing the EMBOSS tools, for example:

• Wageningen Bioinformatics Webportal EMBOSS explorer
• Mobyle@Pasteur
• Soaplab2 Web Services at Vital-IT

For more sophisticated repeat finding you will want to look at tools using Repbase for example:

• CENSOR
  • CENSOR@GIRI
  • CENSOR@EMBL-EBI
• RepeatMasker
• MUMmer (scan_for_match)
• Emboss Palindrome

Other nucleotide repeat finding methods found by a couple of web searches:

• Tandem Repeats Finder
• RECON
• RepeatRunner
• REPuter
• Imperfect Microsatellite Extractor (IMEx)
• Spectral Repeat Finder (SRF)
• REPFIND
• CRISPRfinder
• GrailEXP
• CONREPP
• find_palindromes
• Palindrome
• EMBOSS Palindrome
• Palindrome Search

PANNZER2 processes bacterial proteomes in minutes and eukaryotic proteomes in an hour. You can use AAI-profiler to summarize a proteome's species neighbors and reveal taxonomic identity or contamination.

http://ekhidna2.biocenter.helsinki.fi/sanspanz/

IterPro is for the beginners

https://www.ebi.ac.uk/interpro/

You can find other comparative info at https://academic.oup.com/view-large/118391389

Address of the bookmark: http://ekhidna2.biocenter.helsinki.fi/sanspanz/

http://www.putonti-lab.com/uploads/4/5/3/0/45307835/s-plot2_tutorial.pdf

http://journals.sagepub.com/doi/pdf/10.1177/1176934318797354

Address of the bookmark: https://bitbucket.org/lkalesinskas/splot

SiLiX adopts a graph-theoretical framework to interpret similarity pairs as edges of a network. A very efficient algorithm, based on the Disjoint Sets Data Structure, allows the computation of sequence families with low time and space requirements.

A parallel version of SiLiX, based on MPI, is also available in this package and has been proved to be scalable, so that its allows the study of very large datasets.

SiLiX is already included in the analysis pipeline for HOGENOM.

Address of the bookmark: http://lbbe.univ-lyon1.fr/SiLiX?lang=fr

Address of the bookmark: http://www.littlest.co.uk/software/bioinf/old_packages/miropeats/

NucDiff provides information about the types of differences and their locations. It is possible to upload the results into genome browser for visualization and further inspection. It was written in Python and uses the NUCmer package from MUMmer[1] for sequence comparison.

Address of the bookmark: https://github.com/uio-cels/NucDiff