Induced pluripotent stem cells, or iPSCs, have transformed the way biomedical researchers study human disease. By reprogramming adult cells back into an embryonic-like state, scientists can generate virtually any cell type in the laboratory and watch disease processes unfold in a dish. Yet the reliability of these cellular models depends on a question that has long troubled the field: how faithfully does an iPSC line preserve the genome of the person it came from? A new study published in BMC Genomics by researchers at the Coriell Institute for Medical Research addresses that question head-on, presenting a publicly available iPSC resource in which every line has been scrutinized with long-read whole-genome sequencing, one of the most powerful tools now available for reading the complete human genome.
The resource, described by Laura Scheinfeldt and colleagues, consists of five induced pluripotent stem cell lines derived from apparently healthy donors, each paired with its matched parental cell line. All of the biospecimens are being made available to the scientific community through the National Institute of General Medical Sciences Human Genetic Cell Repository, which is housed at Coriell in Camden, New Jersey. Crucially, the cell lines are accompanied by publicly available high-fidelity, or HiFi, whole-genome sequencing data, along with a suite of web-based search tools that allow any researcher to visualize and explore the genetic variants found in each line.
The choice of sequencing technology matters enormously here. Traditional short-read sequencing, which breaks DNA into fragments of a few hundred base pairs and reassembles them by matching them to a reference genome, is excellent at detecting single nucleotide variants, the single-letter changes in DNA that are the most common form of human genetic variation. But it struggles with structural variants, which include deletions, duplications, inversions, and insertions that can span thousands or even millions of base pairs. These large rearrangements are notoriously difficult to resolve with short reads because the broken pieces of DNA often do not map cleanly to the reference. Long-read sequencing, by contrast, reads individual DNA molecules that stretch tens of thousands of base pairs, allowing structural variants to be observed directly rather than inferred.
HiFi sequencing, the specific long-read approach used in this study, adds another layer of quality. It produces long reads with very high per-base accuracy, combining the reach of long-read technology with the precision traditionally associated with short reads. For a resource intended to serve as a reference standard for the community, that combination is essential. Researchers who download these cell lines and their sequencing data can be confident that the structural variant calls, in particular, reflect genuine features of the genome rather than artifacts of the assembly process.
To assess how well each iPSC line preserved its donor’s genome, the team compared the variants detected in each iPSC line with those in its matched parental cell line. The logic is straightforward: reprogramming should, in principle, leave the genome untouched, so any difference between an iPSC and its parent could signal a change introduced during reprogramming or subsequent cell culture. The results were encouraging overall. Concordance between iPSC lines and their parental counterparts was generally high for both structural variants and single nucleotide variants, suggesting that the reprogramming methods used for most of the lines preserved genomic integrity well.
There was, however, one notable exception, and it carries a cautionary message for the field. One of the five iPSC lines, which had been reprogrammed using a retroviral method, showed a reduction in concordance with its parental cell line. The authors note that this finding is consistent with previously reported concerns about retroviral reprogramming. Retroviruses integrate into the host genome to deliver the reprogramming factors, an approach that has been associated in earlier studies with genomic disruption and with the introduction of copy number changes. While the study does not claim that retroviral reprogramming is inherently unsafe, the observation reinforces a growing consensus that the reprogramming method itself can leave a measurable genomic footprint, and that newer, integration-free methods may better preserve the original genome.
Beyond the concordance analysis, the resource includes annotations that greatly expand its practical usefulness. Each line has been characterized for pharmacogenomic variants, the genetic differences that influence how individuals respond to drugs, and for human leukocyte antigen, or HLA, genes, which encode the proteins that the immune system uses to distinguish self from non-self. HLA typing is critical for any application involving immune compatibility, including potential cell therapies and the development of isogenic controls in immunological experiments. Pharmacogenomic annotation, meanwhile, makes these lines immediately relevant to research on variable drug response, an area where in vitro models derived from well-characterized donors can reveal why the same medication works well for one patient and poorly for another.
The web-based search tools accompanying the resource deserve particular attention because they address a chronic bottleneck in genomics: accessibility of data. Whole-genome sequencing datasets are often deposited in repositories in formats that require substantial computational expertise to parse. By building user-friendly interfaces for visualizing and exploring the structural and single nucleotide variants in these lines, the Coriell team has lowered the barrier for laboratory scientists who may not be bioinformaticians but who need to know, for example, whether a candidate gene in their study carries an interesting variant in one of the available lines. The authors acknowledge the critical contributions of Phillip Hodges to the design, development, and ongoing support of the information technology infrastructure behind these tools, and Jozef Madzo for advice on the genomic data collection.
The samples themselves come from an unusual and forward-looking source. All biospecimens were donated by participants in the Personal Genome Project, an initiative in which volunteers consent to have their biospecimens and associated genomic data used for general research purposes. This open-consent framework means that researchers using these lines can access genomic information that is directly tied to the donor, rather than working with anonymized samples whose genetic data cannot be shared. The repository collection is approved by the Coriell Institutional Review Board, and the participants’ consent covers broad research use, which maximizes the flexibility of the resource for future, unforeseen applications.
The broader significance of this work lies in what it offers to the reproducibility crisis in biomedical research. Cell-based experiments are notoriously sensitive to the identity and quality of the cell lines used, and misidentified or poorly characterized lines have undermined countless studies over the decades. By pairing renewable iPSC lines with matched parental cells, high-quality long-read genome sequences, pharmacogenomic and HLA annotations, and accessible search tools, this resource provides a level of characterization that individual laboratories would find difficult and expensive to replicate on their own. The funding came from the National Institute of General Medical Sciences and the National Human Genome Research Institute, reflecting the federal investment in shared biomedical infrastructure. For researchers interested in the cell-type-specific functional effects of genetic, genomic, and pharmacogenomic variation, the message of the study is clear: a rigorously characterized, openly available set of in vitro models now exists, and it is designed to make the resulting science both higher in quality and easier to reproduce.
Subject of Research: A long-read whole-genome sequencing characterization of induced pluripotent stem cell lines and matched parental cells for enhanced in vitro disease modeling
Article Title: New iPSC resource with long-read whole genome sequencing characterizations for enhanced in vitro modeling
Article References: Scheinfeldt, L., Pompetti, A., Calendo, G., Pozner, T., Grandizio, C., Smith, G., Hodges, K., Gharani, N., Kusic, D., Mitchell, M., & Turan, N. (2026). New iPSC resource with long-read whole genome sequencing characterizations for enhanced in vitro modeling. BMC Genomics. https://doi.org/10.1186/s12864-026-13303-8
Image Credits: AI Generated
DOI: 10.1186/s12864-026-13303-8
Keywords: iPSC, long-read sequencing, HiFi whole-genome sequencing, structural variants, single nucleotide variants, pharmacogenomics, HLA genes, biorepository, reprogramming, in vitro modeling, genomic concordance, Coriell Institute
News Source: Juliet Wilcox. (October 5, 2026). Long-Read Genome Sequencing Powers a New, Deeply Characterized iPSC Resource for Lab Modeling. Scienmag.
