Scimaths.org

Sciences de la Vie et de la Terre · Session 2025 Concours SVT Corrigé détaillé 60 questions corrigées avec explications scientifiques et rappels essentiels. OUISAADAN ABDELHAY   ·   www.scimaths.org Retrouvez les réponses et les justifications des 60 questions du concours, classées par domaine pour faciliter une révision ciblée. Sommaire 01 · Biologie Cellulaire Et Moléculaire (Q1–Q4) 02 · Glucides, Division Cellulaire Et Génétique (Q5–Q14) 03 · Génétique Des Populations Et Immunologie (Q15–Q21) 04 · Physiologie Végétale (Q22–Q27) 05 · Bryophytes (Q28–Q30) 06 · Biologie Et Physiologie Animale (Q31–Q38) 07 · Écologie (Q39–Q45) 08 · Géologie Et Stratigraphie (Q46–Q55) 09 · Géophysique Et Anatexie (Q56–Q60) 01 BIOLOGIE CELLULAIRE ET MOLÉCULAIRE QUESTION 01 Réponse C La structure variable d’un acide aminé est sa chaîne latérale R. Dans la phénylalanine, il s’agit du groupement –CH₂–C₆H₅, repéré C. Le groupe amino, le groupe carboxyle et l’hydrogène porté par le ca...

Corrigé Concours SVT 2024 — OUISAADAN ABDELHAY SVT · SESSION 2024 · CORRIGÉ Imprimer / Enregistrer en PDF SCIENCES DE LA VIE ET DE LA TERRE Concours SVT Corrigé détaillé Session 2024 · Questions 01 à 60 Réponses argumentées et rappels scientifiques Pr. OUISAADAN ABDELHAY www.scimaths.org Ce corrigé accompagne le sujet du concours de spécialité SVT, session d’octobre 2024. Il présente les réponses retenues et leurs explications scientifiques. Les réserves nécessaires sont précisées au fil des questions. GRILLE DES RÉPONSES Cliquez sur une réponse pour accéder au corrigé 01 D 02 D 03 B 04 E 05 A 06 C 07 A 08 A 09 D 10 A 11 C 12 A 13 B 14 A 15 D 16 B 17 D 18 A 19 B 20 B 21 D 22 B 23 E 24 C 25 A 26 C 27 B 28 C 29 D 30 A 31 A 32 C 33 B 34 B 35 E 36 D 37 A 38 A 39 B 40 A 41 A 42 C 43 B 44 C 45 D 46 C 47 C 48 D 49 A 50 B 51 A 52 C 53 C 54 A 55 B 56 B 57 D 58 A 59 D 60 D Biologie cellulaire et moléculaire Génétique Microbiologie et immunologie Biologie et physiologie animale Biologie et physio...

تحميل مقرر تنظيم السنة الدراسية 2026-2027 بالمغرب أصدرت وزارة التربية الوطنية والتعليم الأولي والرياضة المقرر الوزاري رقم 047.28 بتاريخ 03 يوليوز 2026، بشأن تنظيم السنة الدراسية 2026/2027 تحت شعار "من أجل مدرسة ذات جودة للجميع". يهدف هذا المقرر إلى تحديد المحطات والعمليات الأساسية المبرمجة للموسم الدراسي الجديد لضمان تحقيق أهداف خارطة الطريق 2022-2026. أهم تواريخ ومحطات الدخول المدرسي 2026/2027 انطلاق السنة الدراسية: تبدأ السنة الدراسية رسميا يوم الثلاثاء 1 شتنبر 2026. التحاق الإدارة والتفتيش: يلتحق أطر هيئة الإدارة التربوية وهيئة التفتيش بمقرات عملهم يوم الثلاثاء 1 شتنبر 2026. التحاق هيئة التدريس: يلتحق أطر هيئة التربية والتعليم بمقرات عملهم يوم الأربعاء 2 شتنبر 2026. التحاق التلاميذ: يلتحق المتعلمون بشكل تدريجي أيام 3 و4 و5 شتنبر 2026 حسب المستويات الدراسية. الانطلاق الفعلي للدراسة: تنطلق الدراسة بشكل فعلي وإلزامي يوم الإثنين 7 شتنبر 2026 بالنسبة لأسلاك التعليم الأولي والابتدائي والإعدادي والتأهيلي. ...

The Mathematics of LLM Quantization: How to Fit a Massive AI on a Smartphone The rapid evolution of Large Language Models (LLMs) has created a staggering hardware crisis. Modern AI models boast tens or hundreds of billions of parameters. To run a raw, uncompressed 70-billion-parameter model, a computer needs over 140 gigabytes of ultra-fast VRAM—requiring multi-thousand-dollar enterprise GPUs that consume massive amounts of electricity. This hardware constraint makes running advanced AI locally on consumer hardware, like a standard laptop or a smartphone, seemingly impossible. Yet, millions of developers are currently running powerful LLMs locally on everyday hardware. They are achieving this through an elegant mathematical compression technique known as Model Quantization . Here is the linear algebra behind quantization and how it shrinks neural networks without destroying their intelligence. 🔢 1. The Core Problem: The Weight of Floating-Point Math To understand compres...

DNA Data Storage: Why the Future of Cloud Computing is Biological Humanity has a massive, mathematically terrifying problem: we are generating data faster than we can physically build hard drives to store it. As the global datasphere pushes deep into the zettabyte era, relying on traditional silicon flash drives or magnetic tape is becoming unsustainable. We would need to mine massive amounts of rare earth metals and build server farms the size of entire cities, consuming gigawatts of electricity just to keep them cool. To prevent the impending "Data Capacity Crunch," computer scientists have looked away from physics and turned toward biology. They are actively commercializing the most efficient, durable, and compact data storage medium in the known universe: Deoxyribonucleic Acid (DNA) . Here is the incredible science of how tech giants are writing digital code into synthetic biology. 🧬 1. The Mathematics of Extreme Density To understand why DNA is the ultimate...

The "Third Stream" of AI Hardware: How Neuromorphic and Photonic Chips are Saving the Power Grid The artificial intelligence revolution is on a collision course with the global power grid. Traditional data centers powering large language models and generative AI are consuming electricity at an unsustainable rate, with projections showing they could consume up to 3% of the world's total electricity by 2030. The root of the problem isn't the software; it is the physical constraints of the hardware itself. To prevent a massive energy crisis, the semiconductor industry is radically moving away from the traditional silicon designs we have used for decades. In 2026, we are witnessing the explosive commercialization of the "Third Stream" of computing: Neuromorphic Processors and Photonic AI Chips . By mimicking the human brain and computing with actual light, these technologies are delivering mind-bending leaps in energy efficiency and speed. 🧱 1. The...

The "Third Stream" of AI Hardware: How Neuromorphic and Photonic Chips are Saving the Power Grid The artificial intelligence revolution is on a collision course with the global power grid. Traditional data centers powering large language models and generative AI are consuming electricity at an unsustainable rate, with projections showing they could consume up to 3% of the world's total electricity by 2030. The root of the problem isn't the software; it is the physical constraints of the hardware itself. To prevent a massive energy crisis, the semiconductor industry is radically moving away from the traditional silicon designs we have used for decades. In 2026, we are witnessing the explosive commercialization of the "Third Stream" of computing: Neuromorphic Processors and Photonic AI Chips . By mimicking the human brain and computing with actual light, these technologies are delivering mind-bending leaps in energy efficiency and speed. 🧱 1. The...

Physics-Informed Neural Networks (PINNs): Teaching AI the Laws of the Universe For decades, the fields of artificial intelligence and traditional physics operated on entirely different wavelengths. Standard deep learning is purely data-driven: you feed a neural network millions of data points, and it uses statistical patterns to guess an outcome, completely oblivious to the physical rules of the real world. This is why a computer vision model might accidentally generate an image of a coffee cup floating in mid-air—it doesn't understand gravity. In scientific engineering, this blind spot is unacceptable. You cannot rely on "guesses" when designing a bridge, simulating a nuclear reactor, or predicting a hurricane. To bridge this gap, mathematicians and computer scientists have developed a revolutionary paradigm shift: Physics-Informed Neural Networks (PINNs) . By embedding the laws of physics directly into the calculus of machine learning, PINNs are transforming ho...

Crispr-Cas13 and RNA Editing: The New Frontier of Genetic Medicine When CRISPR-Cas9 burst onto the scientific scene, it changed biology forever. It gave scientists the power to act as molecular editors, cutting and pasting sections of DNA to cure genetic disorders. However, DNA editing comes with a major, permanent risk: if the molecular scissors make a mistake—known as an "off-target effect"—that error is permanently written into the patient's genome. To eliminate this risk, researchers have pivoted to a brilliant alternative: CRISPR-Cas13 . Instead of altering DNA, this new system targets and edits RNA . Because RNA molecules are temporary copies of genetic instructions, this breakthrough opens the door to completely reversible, temporary genetic therapies. Here is how Cas13 is rewriting the rules of medicine. 🧬 1. The Central Dogma: DNA vs. RNA To understand why RNA editing is a game-changer, we must look at the software architecture of human cells. Thin...