Transcriptome-Conditioned Personalized De Novo Drug Generation for AML Using Metaheuristic Assembly and Target-Driven Filtering

TL;DR

The study addresses AML’s molecular heterogeneity by proposing an end-to-end framework that ties patient-specific transcriptomics to de novo drug design. It integrates WGCNA-based biomarker prioritization, AlphaFold3 structural modeling, DOGSite pocket profiling, and a reaction-first, fragment-based metaheuristic to generate AML-targeted ligands, with multi-objective optimization guiding synthetic feasibility and binding fit. The approach yields drug-like, synthetically accessible molecules; notably, Ligand L1 achieves a binding energy of $-6.571$ kcal/mol and generated ligands show $QED$ values primarily in the $0.5$–$0.7$ range, indicating pharmacologically viable chemistry. While in silico, the framework demonstrates a scalable blueprint for personalized oncology by bridging transcriptomics, structural biology, and computational chemistry to produce patient-tailored leads for AML and beyond.

Abstract

Acute Myeloid Leukemia (AML) remains a clinical challenge due to its extreme molecular heterogeneity and high relapse rates. While precision medicine has introduced mutation-specific therapies, many patients still lack effective, personalized options. This paper presents a novel, end-to-end computational framework that bridges the gap between patient-specific transcriptomics and de novo drug discovery. By analyzing bulk RNA sequencing data from the TCGA-LAML cohort, the study utilized Weighted Gene Co-expression Network Analysis (WGCNA) to prioritize 20 high-value biomarkers, including metabolic transporters like HK3 and immune-modulatory receptors such as SIGLEC9. The physical structures of these targets were modeled using AlphaFold3, and druggable hotspots were quantitatively mapped via the DOGSiteScorer engine. Then developed a novel, reaction-first evolutionary metaheuristic algorithm as well as multi-objective optimization programming that assembles novel ligands from fragment libraries, guided by spatial alignment to these identified hotspots. The generative model produced structurally unique chemical entities with a strong bias toward drug-like space, as evidenced by QED scores peaking between 0.5 and 0.7. Validation through ADMET profiling and SwissDock molecular docking identified high-confidence candidates, such as Ligand L1, which achieved a binding free energy of -6.571 kcal/mol against the A08A96 biomarker. These results demonstrate that integrating systems biology with metaheuristic molecular assembly can produce pharmacologically viable, patient tailored leads, offering a scalable blueprint for precision oncology in AML and beyond

Figures (26)

• Figure 1: An end-to-end in silico workflow for identifying surface-expressed biomarkers and optimizing candidate small-molecule inhibitors.
• Figure 2: Library sizes were calculated by summing read counts per sample; which the histogram displays a Gaussian-like distribution centered around the mean count depth. The absence of heavy tails indicates no failed libraries or significant outliers in sequence
• Figure 3: The probability density function of gene expression was estimated using Kernel Density Estimation (KDE); which is the plot shows highly consistent expression profiles across all samples (overlapping curves)
• Figure 4: Expression values of the top HVGs were Z-score normalized by gene and clustered using Euclidean distance and Ward’s linkage.
• Figure 5: Sample-Sample Correlation Matrix