Background: Limitations of Conventional Phylogenomic Approaches and the Extreme Homozygosity Genetic Data Bottleneck in Rare Disease R&D

Conventional, linear, and static genomic analysis guidelines have inherent critical blind spots, failing to control for post-mortem DNA damage (deamination) noise and false-positive mutation signals arising from cellular disaggregation and structural decay, which are characteristic of ancient DNA (aDNA). In particular, the prevailing hypothesis that the primary cause of Neanderthal extinction was the accumulation of deleterious recessive mutations due to inbreeding and the resulting collapse of biological fitness has been limited by the inability to fine-tune in silico computational simulations of in vivo effective prophylactic concentrations and the rate of fitness decline. By relying solely on baseline models of large populations, it has failed to predict the genetic gradients and selective pressure changes in small, isolated populations, ultimately failing to systematically interpret the viability flux data in bottleneck microenvironments, leading to significant data bottlenecks and barriers in the discovery of therapeutic targets for rare genetic diseases and the assessment of genetic burden. To overcome this, an integrated computational protocol spanning the entire omics matrix is essential.

Discovery: Implementation of a Computational Omics Pipeline and Demonstration of Population-Scale Resolution Allelic Independent Variable Tensor Synchronization

This study utilized more than 20 new Neanderthal paleogenomic genomes to precisely implement a computational omics pipeline and a batch effect removal technique based on molecular binding free energy calculations. By performing ultra-high-resolution calculations that surpass existing simple analysis models, it demonstrated that even in extreme population bottleneck conditions, the genetic purging mechanism, driven by strong selective pressure, was highly active. By proactively calculating differential equation-based rate constants in silico and identifying the topological variation curves of downstream transcriptomic networks, it was demonstrated that they maintained robust cellular homeostasis until just before extinction, without genetically collapsing completely. This was precisely demonstrated through an allelic independent variable tensor synchronization technique, which represents the organic flow of allelic frequencies at each locus, and precisely demonstrates the molecular biological integrity that directly refutes the prevailing biological dogma in the academic community that long-term inbreeding inevitably leads to fitness collapse and extinction.

Establishment of a Model for Coordinating the Accumulation of Deleterious Alleles and Precisely Stratifying Reversible Homeostasis

The research team implemented a dynamic stratification architecture that analyzes the selective elimination rate of deleterious mutations by virtually tuning the rate-limiting step constants in the metabolic pathways essential for viability. Based on a high-precision omics matrix, a precise stratification model was established based on the frequency gradient of residual deleterious genes within the population, and the ability to maintain reversible homeostasis in fluctuating environmental stress conditions was evaluated. By artificially upregulating or downregulating the activity of specific target gene loci, they successfully identified the molecular biological backbone that allows populations to autonomously regulate reversible homeostasis and avoid biochemical collapse, even in situations of maximized inbreeding. This is a remarkable achievement that demonstrates that molecular-level regulatory circuits can be stably maintained even under anomalous environmental loads within the population, and provides a new theoretical basis for the development of precise therapeutic strategies for rare heterozygous diseases.

Prospects: Establishment of a Programmable Comparative Genomics Standard and Implementation of a Next-Generation IND Digital Governance System

The results of this in silico computational platform revolutionize R&D governance by transforming conventional, static, post-hoc, symptomatic system genetics into a multi-dimensional, tensor-based, programmable genomic informatics. This architecture establishes a computational barrier that eliminates batch-to-batch variation, which is prone to occur in the process of high-throughput screening and the development of cell therapies, by linking a genetic gradient correction coefficient system at the high-throughput screening stage, thereby overcoming inter-species genetic distance. Furthermore, the extreme homozygosity population analysis technology meets the technical specifications of companion diagnostics (CDx) and will function as a digital core asset that drastically shortens the timeline for Investigational New Drug (IND) and cGMP commercial launch approvals by regulatory agencies for genetic intractable diseases. Furthermore, we are confident that it will establish a standard for next-generation evolutionary genomics and upgrade the new drug development infrastructure of the entire bio-industry.