Long-Read Sequencing Steps Into the Clinic for Spinal Muscular Atrophy Diagnosis

Spinal muscular atrophy, one of the most devastating inherited neuromuscular disorders of infancy and childhood, may finally have a diagnostic technology capable of matching its genetic complexity. A new clinical study published in Molecular Genetics & Genomic Medicine reports that single-molecule real-time sequencing, a third-generation long-read platform developed by Pacific Biosciences, can diagnose spinal muscular atrophy and screen carriers with a level of precision that conventional short-read methods have never achieved. In a cohort of 771 individuals screened at a Chinese medical center, the long-read approach agreed perfectly with standard quantitative PCR on every sample, while simultaneously delivering information that qPCR and MLPA simply cannot provide: exact copy numbers of both disease-causing and modifier genes, full-length sequencing of a notoriously repetitive genomic region, and the ability to detect point mutations in the same run.

The biology behind spinal muscular atrophy explains why this region has long frustrated diagnosticians. The disease is caused by loss or mutation of the SMN1 gene, which encodes a protein essential for the survival of motor neurons. Humans, however, carry a nearly identical backup copy called SMN2, distinguished by a single nucleotide change at position c.840 in exon 7. That one base difference alters how SMN2 pre-messenger RNA is spliced, so most SMN2 transcripts are translated into a truncated, unstable protein that is rapidly degraded. Each SMN2 copy produces only about ten percent of the functional protein generated by SMN1, and the proportion varies across tissues. Because the two genes are more than ninety-nine percent identical, distinguishing them requires reading across the entire roughly 28.5-kilobase locus, something short-read sequencing struggles to accomplish reliably.

Traditional testing reflects these limitations. Quantitative PCR can detect homozygous deletions of SMN1 exons 7 and 8, the molecular signature of most SMA cases, but it cannot report precise copy numbers or identify point mutations. Multiplex ligation-dependent probe amplification, or MLPA, reliably measures SMN1 and SMN2 copy numbers and detects exon rearrangements, yet it cannot resolve haplotypes or sequence-level variants. Neither method can determine whether a person with two SMN1 copies carries both on one chromosome or one on each, a distinction that matters enormously for reproductive risk. A so-called 2+0 silent carrier, with both copies on a single chromosome, appears normal on standard screening but faces the same one-in-four chance of having an affected child as a classic 1+1 carrier when paired with another carrier.

Single-molecule real-time sequencing sidesteps these obstacles through sheer read length. In the new study, researchers at Zhangzhou Municipal Hospital extracted genomic DNA from peripheral blood, amplified the full-length SMN1 and SMN2 genes together with their downstream regions using long-range PCR, and ligated the amplicons into dumbbell-shaped SMRTbell libraries. These libraries were sequenced on a Sequel II CNDx platform in circular consensus sequencing mode for thirty hours, generating high-fidelity reads that span the entire locus. Because each read covers the whole gene, the c.840 site unambiguously labels each molecule as SMN1 or SMN2, allowing the software to count copies of each gene separately and scan for point mutations at the same time. The technology carries no GC bias and handles the highly repetitive sequence architecture of this region that defeats shorter reads.

The clinical validation was striking. Among the 771 screened individuals, SMRT sequencing and qPCR produced fully concordant results in one hundred percent of samples. The most common genotype was two copies each of SMN1 and SMN2, found in 55.25 percent of participants, followed by a 2:1 ratio in 29.44 percent. Two SMN1 copies were present in 91.18 percent of the cohort, but the distribution revealed important outliers: 2.33 percent of individuals carried only one SMN1 copy and were classified as deletion carriers, yielding a carrier frequency of 2.33 percent, consistent with previously reported rates of 1.2 to 2.2 percent in Chinese populations. More provocatively, 6.36 percent carried three SMN1 copies and one individual carried four, indicating that at least two SMN1 genes sit on a single chromosome in these people, a configuration that standard qPCR cannot distinguish from the benign two-copies-on-two-chromosomes arrangement.

That finding carries direct reproductive consequences. Individuals with elevated SMN1 copy numbers may be silent carriers whose risk is invisible to conventional screening, since qPCR reports only total copy number without allele distribution. The study authors note that definitive assignment of 2+0 status still requires family-based haplotype phasing and pedigree analysis, which lay beyond the scope of the current work. Even so, the results underscore a hidden reservoir of reproductive risk in the general population. With roughly one in fifty people in many populations carrying an SMN1 deletion, and couples who are both carriers facing a one-in-four chance of an affected child in each pregnancy, carrier screening has become a public health priority, and the accuracy of that screening depends entirely on the resolution of the underlying technology.

The study also catalogued the distribution of SMN2, the modifier gene that shapes disease severity. Two copies of SMN2 were the most common configuration at 57.59 percent, one copy accounted for 33.07 percent, and small fractions of individuals carried three or four copies. Notably, 6.10 percent of the screened population carried homozygous SMN2 deletions, a genotype that is not itself disease-causing but eliminates the compensatory buffer that SMN2 provides. Because SMN2 copy number is inversely related to clinical severity in SMA patients, knowing a patient’s SMN2 count helps predict disease course and informs treatment decisions, particularly now that disease-modifying therapies exist for a condition that was, until recently, essentially untreatable.

Four qPCR-confirmed SMA patients served as independent validation cases, and their stories illustrate both the technology’s power and the disease’s variability. Patient 1, a 16-year-old female with severe scoliosis who first showed hand tremors at twelve months, had zero copies of SMN1 and three copies of SMN2 confirmed by all three methods, consistent with SMA type II. Patient 2, a 12-year-old male unable to walk independently at seventeen months, showed the same molecular profile with profound proximal weakness and absent reflexes, also diagnosed as type II. Patient 3, a 17-year-old male with flexion contractures, generalized hypotonia, and complete lower limb paralysis, carried zero SMN1 copies and three SMN2 copies but was classified as type I. Patient 4, a three-month-old male with hypotonia since birth, had zero SMN1 copies and two SMN2 copies, with electromyography revealing nerve fiber damage across multiple limbs, confirming type I disease. No pathogenic point mutations were detected in any patient.

The discordance among patients sharing three SMN2 copies is scientifically telling. Three of the four patients carried the same SMN2 count yet spanned two clinical types, echoing a large analysis of 1,627 SMA patients with three SMN2 copies in which fifteen percent had type I disease, fifty-four percent type II, and thirty-one percent type III. The relationship between copy number and severity is clearly not a simple inverse correlation; even among patients with four or five SMN2 copies, one and five percent respectively still develop type I disease. Whether different SMN2 variants possess distinct functional properties, or whether modifying mutations within SMN2 influence its compensatory capacity, remains an open question that the authors flag for future investigation.

The study’s limitations are candidly acknowledged. Point mutations in SMN1 account for roughly two percent of SMA carrier alleles, yet none were detected in this cohort, a result that may reflect their genuine rarity or a sample size insufficient to capture such infrequent events. Larger-scale studies will be needed to validate the technology’s performance on rare genotypes. Nevertheless, the trajectory is clear: SMA genetic testing has evolved from single-site detection toward whole-gene, haplotype-level analysis, and long-read sequencing has already matured into clinical practice for other monogenic disorders such as thalassemia and congenital adrenal hyperplasia. With its ability to resolve copy numbers, haplotypes, and point mutations in a single assay across one of the human genome’s most treacherous paralogous regions, single-molecule real-time sequencing appears poised to become a cornerstone of SMA diagnosis, carrier screening, and genetic counseling, potentially catching silent carriers that today’s methods routinely miss.

Subject of Research: Clinical application of single-molecule real-time long-read sequencing for spinal muscular atrophy diagnosis and carrier screening

Article Title: Clinical Practice of Single‐Molecule Real‐Time Sequencing in the Diagnosis of Spinal Muscular Atrophy

Article References: Wei, H., Zheng, Y.-W., Zhang, N.-K., Gao, T.-L., Zheng, J.-X., Li, T.-Y., & Hong, S.-R. (2026). Clinical Practice of Single‐Molecule Real‐Time Sequencing in the Diagnosis of Spinal Muscular Atrophy. Molecular Genetics & Genomic Medicine, 14(10), Article e70314. https://doi.org/10.1002/mgg3.70314

Image Credits: AI Generated

DOI: 10.1002/mgg3.70314

Keywords: spinal muscular atrophy, SMRT sequencing, SMN1, SMN2, long-read sequencing, carrier screening, genetic diagnosis, copy number variation, qPCR, MLPA, motor neuron disease, genetic counseling