Answers are grounded in this paper and cite their sources.
Colorectal cancer (CRC) is a molecularly heterogeneous malignancy requiring reproducible biomarkers and structurally tractable therapeutic targets. We applied an integrated transcriptomic, network, virtual-screening, docking, ADMET, and molecular-dynamics (MD) workflow using GSE8671, GSE14333, and GSE89076. Differential expression analysis (adjusted p < 0.05, |log2FC| ≥ 1) identified 2,092 upregulated/9130 downregulated genes in GSE8671, 926/945 in GSE14333, and 2,317/1,456 in GSE89076. Cross-dataset concordance was highest between GSE8671 and GSE89076 ( r = 0.70), with 1,040 shared upregulated and 500 shared downregulated genes, whereas GSE14333 showed weak concordance with GSE8671 ( r = −0.11) and GSE89076 ( r = −0.01). Integrated network topology, enrichment, and expression profiling prioritized SMAD5, USP2, and CUL4A. Structure-based virtual screening of approximately 25,000 approved/bioactive compounds identified Metergotamine–SMAD5 (−10.0 kcal/mol), Ravidasvir–USP2 (−12.5 kcal/mol), and Hesperidin–CUL4A (−10.8 kcal/mol) as leading screening hits. Refined molecular docking subsequently prioritized MK3207–SMAD5 (−10.6 kcal/mol), R428–USP2 (−12.1 kcal/mol), and Hesperidin–CUL4A (−8.7 kcal/mol). ADMET profiling indicated the most balanced drug-likeness for MK3207, poor predicted solubility for R428, and multiple rule-based violations for Hesperidin. Each complex underwent a 100 ns all-atom MD simulation in explicit solvent. CUL4A–Hesperidin retained hydrogen bonds with GLU680, THR537, GLN673, and THR535 in >83% of frames and a LYS459 π-cation interaction in 73.3%. USP2–R428 maintained key PHE364, SER190, and ARG363 contacts with a protein-referenced ligand RMSD of 4.84 ± 0.58 Å. Conversely, SMAD5–MK3207 showed substantial displacement (14.09 ± 0.62 Å) and <4% docking hydrogen-bond occupancy. Single-frame 100 ns MM-GBSA endpoint energies were −107.62, −95.99, and −88.00 kcal/mol, respectively. Collectively, the results support CUL4A–Hesperidin and USP2–R428 as the most dynamically persistent complexes, while the initial SMAD5–MK3207 binding mode requires re-evaluation and experimental validation.
No takes yet. Share an insight, caveat, or question.
Alqarni et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: