Talos, an open-source tool, automates reanalysis of genomic data to speed rare disease diagnosis

Genomic testing has transformed the diagnosis of rare disease, but more than half of patients remain undiagnosed after their first test. Because our understanding of the genome improves constantly, stored sequencing data can be reexamined to yield diagnoses that were impossible to make initially. However, reanalysis today is overwhelmingly manual, relying on motivated clinicians and scarce laboratory staff, so the vast majority of stored genomes are never revisited.

Talos, an open-source tool developed through a collaboration spanning the Centre for Population Genomics, Australian Genomics, the Broad Institute, and Microsoft, was designed to automate this process. It reinterprets a patient's existing variant calls against the latest community knowledge each time it runs, drawing on two continuously updated public resources: PanelApp Australia for gene-disease relationships and ClinVar for variant-level pathogenicity. The tool is optimized for a low false-positive rate, returning a small set of high-confidence variants rather than a long ranked list, because in real-world genomic reanalysis the limiting factor is human review time.

Validated against expert manual analysis

Talos was benchmarked on two independent cohorts that had undergone careful manual analysis: the Australian Acute Care Genomics (ACG) cohort of critically ill infants and children, and the U.S.-based Rare Genomes Project (RGP) cohort. Across 1,089 probands, Talos recovered 90% of in-scope diagnoses on the ACG cohort while returning a median of just 1.3 candidate variants per family. On the RGP cohort, it recovered 87% of in-scope diagnoses at the same median rate, showing generalizability.

In a head-to-head comparison with Exomiser, a widely used prioritization tool, Talos matched its overall sensitivity for small variants but operated at a very different point: Exomiser returns a broad ranked list, while Talos returns a short, highly specific one. When review was limited to a realistic budget of the top five or top one ranked variants, Talos came out significantly ahead (p = 0.017 and p < 0.0001, respectively). The two tools surfaced different variants, suggesting they are complementary and should ideally be used together in diagnostic workflows.

Deployed on an international scale

The most significant experiment involved a tested-but-undiagnosed cohort of 4,735 individuals drawn from Australian Genomics research studies and a single diagnostic laboratory. Talos produced 241 new diagnoses in 238 individuals, a 5.1% additional yield, with every single likely-causative variant subsequently confirmed as pathogenic or likely pathogenic by accredited labs.

The sources of those diagnoses illustrate why reanalysis is such a powerful paradigm: 32% came from new gene-disease relationships discovered since the original test, 22% from new variant-level evidence (reclassifications), and 45% from improved filtering and analysis, including variant types such as copy number variants and structural variants not examined originally, and phenotype filters that had been set too narrowly. Yield was consistent across clinical areas (roughly 5-6% for neurodevelopmental, cardiac, and renal indications), though the reasons differed.

Genome data outperformed exome data (6.1% vs 4.8%), partly by reaching non-coding diagnoses such as RNU4-2 and a deep-intronic MRPL39 variant. Notably, 59% of the new gene-disease diagnoses were not yet curated in OMIM at the time of reanalysis, underscoring the value of drawing on a rapidly updated resource like PanelApp Australia.

From a one-off event to a continuous program

Talos was then run for 29 monthly iterative cycles. While most diagnoses (92%) came on a cohort's first pass, the iterative design proved its value on two fronts. First, because later cycles return only newly actionable evidence, they surfaced an average of just one variant per 200 cases over the program, demonstrating scalability. Second, the tool showed how quickly we can move from scientific discovery to diagnosis: on average just 32 days passed between new knowledge appearing in a public database and a patient receiving a diagnosis, with the fastest case turning around in a single day.

The pipeline is cheap enough to run continuously: annotating 1,000 genomes cost about $11, and a monthly reanalysis pass ran for a few cents per cohort.

Looking ahead

Talos reframes genomic reanalysis from a rare, labor-intensive event into a continuous, automated program that can keep pace with the science. By optimizing for specificity, it respects the real bottleneck of expert reviewer time, and by drawing on openly shared, frequently updated resources, it turns the global community's accumulating knowledge into diagnoses for individual patients, often within weeks.

The developers believe they have established a foundational capability and are excited to see how the community builds on it. In particular, as more advanced AI models for understanding and predicting the consequences of genetic variation become available, they look forward to leveraging them in the reanalysis of unsolved rare disease cases.

Talos is open source and straightforward to deploy in cloud environments like Azure. The results offer a practical blueprint for health systems aiming to deliver frequent, scalable reanalysis to the many patients still searching for diagnoses.