BEGIN:VCALENDAR VERSION:2.0 PRODID:-//CERN//INDICO//EN BEGIN:VEVENT SUMMARY:Thermomechanical Mapping of DNA under Coupled Salt-Temperature Con trol with oxDNA2 Coarse-Grained Simulations DTSTART;VALUE=DATE-TIME:20260327T111000Z DTEND;VALUE=DATE-TIME:20260327T113000Z DTSTAMP;VALUE=DATE-TIME:20260426T035628Z UID:indico-contribution-806-10343@events.saip.org.za DESCRIPTION:Speakers: Isaiah Igwe (Federal University Dutsin-Ma)\nQuantify ing how temperature and ionic strength jointly determine the elastic prope rties of double-stranded DNA remains a central challenge in molecular biop hysics. Although individual temperature or salt dependent trends have been measured\, a unified\, mechanistic map across high-salt and elevated-temp erature regimes is still lacking. Here\, we use the oxDNA2 coarse-grained model to compute a 3×9 thermodynamic ionic grid spanning 300 –373 K and 0.5 – 1.5 M monovalent salt\, enabling controlled evaluation of elastic properties and structural observables under conditions where electrostati c screening is extremely strong. Simulations were performed on long 500-bp duplexes with full ensemble averaging over independent replicate trajecto ries at each condition. Across all salt concentrations\, DNA softens as te mperature increases\, but the degree of softening depends on ionic strengt h. At 0.5 M\, the bending persistence length drops from about 43 nm at 300 K to 32 nm at 373 K. At 1.5 M\, the decrease is far smaller (46 → 39 nm )\, showing that high salt reduces thermal sensitivity by roughly 40–50% . Torsional stiffness shows the same pattern (110 → 92 units at 0.5 M vs . 118 → 105 units at 1.5 M)\, as does twist–stretch coupling\, which c hanges by 0.7 units at low salt but only 0.4 units at high salt. Helical t wist decreases by roughly 1.1–1.3° per kbp per 10 K at 0.5 M\, with a v isibly weaker dependence at 1.5 M. While both bending and torsional rigidi ties soften with temperature\, torsional elasticity remains closer to harm onic behavior than bending\, with anharmonic deviations staying below ~10% even at the highest temperatures. Structural measures\, including base-pa ir occupancy and stacking energies\, show that the duplex remains intact u p to about 95 –100 °C\, with only limited end fraying. Beyond quantifyi ng these trends\, the present work introduces a unified thermodynamic inte rpretation of DNA elasticity. The simulations demonstrate that the weakeni ng of base-stacking interactions precedes significant hydrogen-bond disrup tion and acts as the primary microscopic driver of thermoelastic softening . This perspective provides a compact statistical-mechanical description o f DNA thermoelasticity and helps reconcile observations from coarse-graine d simulations\, atomistic molecular dynamics\, and single-molecule experim ents.\n\nhttps://events.saip.org.za/event/272/contributions/10343/ LOCATION: URL:https://events.saip.org.za/event/272/contributions/10343/ END:VEVENT BEGIN:VEVENT SUMMARY:Integrating Generative AI\, Computational Modeling\, and Physiolog ical Reasoning to Enhance Biological Sciences Education DTSTART;VALUE=DATE-TIME:20260327T105000Z DTEND;VALUE=DATE-TIME:20260327T111000Z DTSTAMP;VALUE=DATE-TIME:20260426T035628Z UID:indico-contribution-806-10338@events.saip.org.za DESCRIPTION:Speakers: Camellia Okpodu (University of Wyoming)\nPreparing f uture biophysicists requires approaches that connect foundational biologic al principles with modern computational tools. Our work introduces a Gener ative‑AI–enhanced framework for teaching bioinformatics\, designed to strengthen students’ computational reasoning\, data literacy\, and engag ement with cardiovascular‑related biological systems. This model emphasi zes ethical and effective integration of AI outputs into analysis and mode ling\, helping learners navigate emerging digital research environments. Building on this framework\, we developed a MATLAB‑based SpO₂ modeling exercise that guides students through finite‑difference modeling of oxy gen transport\, clinical decision‑making\, and the interpretation of phy siological data. By incorporating AI‑generated clinical scenarios into M ATLAB workflows\, students explore realistic diagnostic pathways and deepe n understanding of physiological mechanisms. Together\, these innovations create an accessible instructional pipeline—particularly valuable for st udents across the African diaspora—linking computational physiology\, ca rdiovascular innovation\, and AI‑supported reasoning. This combined appr oach broadens participation in biophysics education and offers scalable mo dels for strengthening quantitative and computational skills in the biolog ical sciences.\n\nhttps://events.saip.org.za/event/272/contributions/10338 / LOCATION: URL:https://events.saip.org.za/event/272/contributions/10338/ END:VEVENT BEGIN:VEVENT SUMMARY:Steroidal Pregnanes as 11β-HSD1 Modulators: Insights from Random Forest-Based QSAR and Atomistic Simulations DTSTART;VALUE=DATE-TIME:20260327T103000Z DTEND;VALUE=DATE-TIME:20260327T105000Z DTSTAMP;VALUE=DATE-TIME:20260426T035628Z UID:indico-contribution-806-10309@events.saip.org.za DESCRIPTION:Speakers: Oludare Ogunyemi (University of Ibadan)\nThe enzyme 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) is a validated therap eutic target for Type 2 diabetes mellitus due to its role in local regener ation of active glucocorticoids. Current inhibitors are often limited by t heir constrained chemical diversity\, moderate potency\, and off-target ef fects. The therapeutic potential of steroidal pregnanes in diabetes and me tabolic disorders is been widely reported\; however\, their molecular targ ets\, particularly in glucocorticoid signaling\, remain poorly understood. This study explored the interactions of a curated library of steroidal pr egnanes with 11β-HSD1 using integrated Machine Learning (ML)-based QSAR\, molecular docking\, 100 ns Molecular Dynamics (MD) simulations\, and MM-G BSA binding free energy calculations. Initial exploratory chemical space a nalysis of the IC50 bioactivity dataset revealed that hydrogen donors\, mo lecular weight\, and lipophilicity may contribute to the bioactivity of 11 β-HSD1 inhibitors. Evaluation of 42 ML algorithms based on performance me trics revealed Random Forest Regressor (RFR) as a top model for bioactivit y predictions. Molecular docking simulation of the top RFR-predicted comp ounds (pIC50 ≥ 6.0 and pKi ≥ 7.8) with the active site of 11β-HSD1 id entified three compounds (pregnane-3\, 20-diol disulphate (P1)\, 20-Piperi din-2-yl-5α-pregnan-3β\,20-diol (P2)\, and 12\,20-di-O-benzoyl-pregnane- 3β\,12β\,14β\,20-tetraol (P3)). While the reference carbenoxolone prima rily strongly involved peripheral polar contacts to stabilize its orientat ion\, the pregnane scaffolds demonstrated deeper insertion into the hydrop hobic catalytic cavity of 11β-HSD1\, resulting in enhanced shape compleme ntarity and van der Waals packing. The thermodynamic parameters computed f rom the MD simulation trajectories revealed both the structural stability and intrinsic conformational flexibility of the 11β-HSD1–pregnane compl exes. Moreover\, the lower MM-GBSA binding energies of P1 (-43.58 kcal/mol ) and P3 (-44.95 kcal/mol) as compared with the reference carbenoxolone (- 24.19 kcal/mol) indicate high binding affinity and validate the docking sc ores of the hits. Additionally\, the leads exhibited favorable physicochem ical and pharmacokinetic profiles. Overall\, our findings provide mechanis tic insights into ligand binding and highlight key structural features tha t may account for 11β-HSD1 modulation by steroidal pregnanes\, offering a framework for the rational design of pregnane-derived therapeutics.\n\nht tps://events.saip.org.za/event/272/contributions/10309/ LOCATION: URL:https://events.saip.org.za/event/272/contributions/10309/ END:VEVENT END:VCALENDAR