Biochemistry Question Quality Review

Executive Summary

The 200-question candidate sample for Biochemistry reveals a subject pool with a structurally sound factual foundation but a severe and consistent quality gap relative to the benchmark and PYQ standards. The central problem is not isolated bad questions — it is a systemic overrepresentation of bare-recall, single-sentence items that test definitional memory rather than clinical reasoning or mechanistic understanding.

Across all eight shards, approximately 60–65% of candidate questions operate at Bloom's Level 1 or 2 with no clinical context, no vignette framing, and no meaningful distractor challenge. The benchmark set, by contrast, operates uniformly at Bloom's Level 3 with patient scenarios, integrated lab values, and distractors that require genuine discrimination. The PYQ set includes some Bloom's 1–2 items but these are anchored to high-yield, exam-validated concepts and are not the dominant format.

Beyond the recall overload, the reviewed set contains a non-trivial rate of outright factual errors in answer keys — a harder problem than low Bloom's level because these items actively misinform learners. A third distinct problem is structural: questions with duplicate options, missing images, incoherent distractor sets, and "All of the Above" keys that reward test-taking strategy over knowledge. Finally, a smaller but operationally important cluster of questions tests fringe, obsolete, or misclassified content that has no place in an INI-CET/NEET-PG preparation pool.

The good news is that a meaningful minority of candidate questions — estimated at 25–30% of the sample — have sound factual cores and could be upgraded to benchmark quality with targeted editorial work. The subject is not broken; it is undertreated.

Summary disposition across the 200-question sample (estimated from shard evidence):

Disposition	Estimated Count	Rationale
Keep as-is	~35–40	Clinically framed, factually sound, appropriate distractors
Fix (content or structure)	~55–65	Sound concept, fixable error or upgradeable stem
Disable	~95–110	Recall-only trivia, factual errors, structural failures, fringe content

What Good Looks Like

The benchmark and PYQ sets define the quality bar clearly and consistently. Good questions in this subject share five properties:

1. A clinical or physiological anchor. Every benchmark question opens with a patient scenario that provides age, presentation, and relevant investigations. The question does not ask "what enzyme is deficient in Hurler syndrome?" — it presents a child with organomegaly, corneal clouding, and urinary mucopolysaccharides and asks the examinee to identify the deficient enzyme from that evidence. The reasoning chain is the point, not the label.

2. A single, unambiguously correct answer. The correct option is defensible against all four major Indian PG biochemistry references (Harper's, Lehninger, Stryer, Vasudevan). Distractors are plausible — they represent common confusions or related enzymes/pathways — but are clearly distinguishable by a well-prepared candidate. No option is a near-synonym of the key, and "All of the Above" is never the correct answer.

3. Bloom's Level 3 or higher as the default. Even benchmark questions tagged "easy" require application: the Wernicke's encephalopathy question (f4261145) is conceptually simple but demands that the candidate apply the rule "give thiamine before dextrose" to a specific clinical scenario. The Lesch-Nyhan question (b9c5c07c) requires the candidate to read a family history and infer X-linked recessive inheritance, not just recall the enzyme name.

4. High-yield topic selection. Benchmark and PYQ items cluster around topics with demonstrated exam frequency: urea cycle disorders, lysosomal storage diseases, vitamin deficiencies with clinical presentations, enzyme kinetics, lipid metabolism disorders, and molecular biology techniques. Obscure colorimetric tests, obsolete unit systems, and food composition trivia are absent.

5. Factual precision in both stem and options. The benchmark questions are written with enough specificity that the correct answer is not contestable. The CPS-I deficiency question (1ebafee0) specifies plasma amino acid patterns (low citrulline, low arginine) that distinguish CPS-I from OTC deficiency — a level of precision that makes the question genuinely discriminating rather than pattern-matched.

The PYQ set adds one additional quality marker: exam pedigree. Questions like the non-competitive inhibition item (6d1764d7) and the heme catabolism "NOT a product" item (c2b4fb5e) have been validated by actual exam performance. They are not always at Bloom's 3, but they test concepts that have appeared repeatedly across INICET and NEET-PG cycles, confirming their relevance.

Main Issue Categories

1. Recall-Only Bare-Stem Questions: The Dominant Volume Problem

Why this pattern is bad

This is the single largest quality problem in the reviewed set, appearing across every shard without exception. Questions that ask "Which metal does ceruloplasmin contain?" or "Transcription is the synthesis of which molecule?" or "Melatonin is synthesized from which amino acid?" do not discriminate between a well-prepared PG candidate and a first-year MBBS student who has read one page of a textbook. They add no diagnostic value to a practice test, they do not build clinical reasoning, and they crowd out space that could be occupied by higher-quality items. The Bloom's distribution data confirms the scale: 96 of 200 candidate questions are at Bloom's Level 1, and the actual cognitive demand of many Bloom's 2-tagged items is closer to Level 1 in practice.

How it shows up

The pattern appears as one-sentence stems with a single factual answer and distractors that are obviously wrong to any prepared candidate. The clinical context is entirely absent. Topics affected span the entire subject: carbohydrate metabolism pathway steps, amino acid precursor relationships, vitamin deficiency single-sign associations, enzyme cofactor identities, and molecular biology definitions.

Example question IDs

cd56e8d9 — "Ceruloplasmin contains which metal?" No clinical context, Bloom's 1, zero discriminatory value at PG level.
6eb16a69 — "How many Fe²⁺ atoms in one haemoglobin molecule?" Pure numerical recall, no reasoning required.
97c8426a — "Transcription is the process of synthesis of RNA." This is a textbook definition restated as a question. No candidate who has reached PG entrance preparation should miss this.
542fb4b4 — "Melatonin is synthesized from which amino acid?" Single-fact recall, Bloom's 1, no clinical hook.
9dcc072c — "Tertiary protein structure definition." Verbatim textbook definition as a question stem.
59b59de5 — "Promoter = cis-acting factor." Definitional recall with no application.
5c78b785 — "Which enzyme does not catalyze a reversible step in glycolysis?" List-recall of irreversible glycolytic steps, no clinical framing.
44c605a3 — "SDS-PAGE separates proteins by size." Trivial single-fact recall.
a80ec08d — "Karyotyping is useful for diagnosing chromosomal abnormalities." Self-evident to any medical student.
04d14e00 — "Hemochromatosis is a defect in the metabolism of which element?" The most basic possible recall item on this topic.
58b40a07 — Glycosidic linkage definition. No application.
55dfa54a — Epimer of glucose. Single-word answer, Bloom's 1.

Recommended disposition

Disable the pure recall items where strong gold-standard coverage of the same concept already exists at higher Bloom's levels in the benchmark or PYQ sets. For items where the underlying concept is high-yield but the question is bare-stem (e.g., thiamine cofactor role, GLUT transporter specificity, enzyme kinetics), flag for vignette upgrade rather than outright disabling — these are Fix candidates, not Disable candidates. The distinction matters operationally: disabling removes the item; fixing requires editorial resource allocation.

2. Factual Errors and Wrong Answer Keys: The Safety-Critical Problem

Why this pattern is bad

This is the highest-severity issue in the reviewed set because it actively misinforms learners. A student who studies from a question with a wrong answer key will encode incorrect information and may carry that error into the actual examination. Unlike low Bloom's level — which is a quality gap — a wrong answer key is a patient safety concern in the long run and an immediate credibility risk for the platform. The rate of factual errors observed across shards is non-trivial: at least 8–10 confirmed or strongly suspected wrong keys were identified in 200 questions, suggesting a pool-level error rate that warrants a systematic audit.

How it shows up

Wrong keys appear in two forms. The first is an outright factual inversion: the marked correct answer is the opposite of what standard references state. The second is a question where multiple options are simultaneously correct, making the marked key defensible but not uniquely correct.

Example question IDs

f073adfe — Net ATP yield from glycolysis marked as 7. Standard biochemistry: net ATP from glycolysis = 2 (anaerobic). The answer 7 is incorrect by any conventional framework. This is a straightforward factual error.
762bd09e — "Ascorbic acid decreases iron absorption" marked as correct. Ascorbic acid enhances non-heme iron absorption by reducing Fe³⁺ to Fe²⁺. The correct inhibitor is phytate. This is a factual inversion.
108889f2 — "Albumin" marked as the mechanism of intramitochondrial fatty acid transport. Albumin is a plasma carrier; the mitochondrial transport mechanism is the carnitine shuttle. Wrong answer key.
b190c2ad — "Aspartate" marked as the substrate from which urea is formed. Urea is cleaved from arginine by arginase; aspartate donates a nitrogen but is not the direct substrate. Requires correction to "Arginine" or a rewritten stem.
3ac8afe1 — "Protein" marked as the only macromolecule requiring a template for synthesis. DNA and RNA also require templates. The question is factually incorrect as framed.
d3cc71cf — Xeroderma pigmentosum attributed to "DNA polymerase I and DNA ligase." XP involves nucleotide excision repair endonucleases (XPC, XPA, ERCC2). This is a clear factual error.
be2dd36d — "Pyruvate" marked as the first substrate of the Krebs cycle. The cycle begins with condensation of acetyl-CoA and oxaloacetate; pyruvate is the precursor to acetyl-CoA, not a cycle substrate.
c76cd96d — Gluconeogenesis question where option A ("occurs primarily in the liver") is a correct statement by standard teaching, yet a different option is marked as the answer, creating an ambiguous or erroneous key.
866a9a94 — Immune complex formation attributed to quaternary protein structure. Antigen-antibody binding involves tertiary surface complementarity, not quaternary structure. Risks teaching an incorrect concept.
0deb82fc — "Hormone not acting through cAMP" question where both progesterone and estrogen (steroid hormones acting via nuclear receptors) are listed as distractors alongside GH, making multiple options simultaneously correct.

Recommended disposition

All confirmed factual errors must be disabled immediately pending correction. Items with ambiguous keys (multiple simultaneously correct options) should be fixed by rewriting the stem to make only one option unambiguously correct, or by replacing the problematic distractors. No item with a wrong answer key should remain active in any test template. This category requires a dedicated expert review pass across the full 10,646-question pool, not just the sampled set — the error rate in the sample suggests the problem is not isolated.

3. Structural MCQ Construction Failures: Format Problems Independent of Content

Why this pattern is bad

This category is distinct from factual errors because the content may be correct but the question format undermines its validity as an assessment tool. "All of the Above" as the correct answer rewards elimination strategy over knowledge — a candidate who is certain of any two options can select "All of the Above" without knowing the third. Missing images make questions unanswerable. Duplicate options within a single question are authoring errors that signal the item was never properly reviewed. Near-synonym distractors (where two options say essentially the same thing with different wording) reduce the effective choice set and inflate guessing probability. These are not content problems; they are construction problems that require a different remediation path.

How it shows up

The pattern appears as "All of the Above" keys, questions referencing pedigrees or figures that are not embedded in the item, duplicate option text within a single question, and distractor sets where options belong to a different ontological category than the correct answer (e.g., amino acids listed as distractors in a question about enzymes).

Example question IDs

7a6eb487 — "All of the Above" as the correct answer for an isopeptide bond question. Rewards elimination, not knowledge.
e1359dbb — "All of the Above" as the correct answer for an isoenzyme question.
62cd585c — References a pedigree image that is absent from the text. The question is unanswerable without the image. Additionally uses "All of the Above" as the correct answer.
60cc36e7 — "All of the Above" as the correct answer for a DNA structure question.
298aeaeb — "All of the Above" as the correct answer for a lysosomal storage disorder question.
9d940927 — Option A and option D are both printed as "1.5 ATP" — a duplicate option error that reduces the effective choice set to three.
340440db — "Folic acid" appears as both the correct answer (option A) and a distractor (option D) in the same question. Additionally, the stem asks for "an amino acid" but folic acid is a vitamin.
3c8b2a11 — Chaperone function question where three of the four options are amino acids (valine, threonine, aspartate) rather than proteins or processes — the distractor set is ontologically incoherent.
8a8fc1ca — Endonuclease function question where distractors ("Enhancers," "identify antibiotic resistances," "identify coding regions") are not plausible enzymatic alternatives.
34c06835 — "Termination mutation" listed as a distractor for a question where "Nonsense mutation" is the correct answer; the two terms are near-synonyms in standard usage, creating genuine ambiguity.
102648d4 — Stem references "X" and "Y" in a matching format but no table or figure is provided; the question is a fragment.

Recommended disposition

All questions with missing images should be disabled immediately — they are unanswerable as delivered. Questions with duplicate options or ontologically incoherent distractor sets should be fixed before any further use. "All of the Above" items should be restructured as single-best-answer questions; if the concept cannot be tested without "All of the Above," the question should be rewritten from scratch. Near-synonym distractor pairs should be resolved by replacing one with a clearly distinct alternative.

4. Thin Clinical Vignettes That Resolve at Recall Level

Why this pattern is bad

This category is distinct from pure recall (Category 1) and distinct from high-quality clinical vignettes. It describes questions that have a clinical shell — a patient age, a symptom, sometimes a lab value — but the clinical information is decorative rather than functional. The correct answer can be retrieved by pattern-matching a textbook triad without actually processing the clinical scenario. These questions are not as harmful as wrong-key items, but they represent a missed opportunity: the editorial investment of writing a vignette was made, but the reasoning demand was not elevated to match. They are the most fixable category in the pool.

How it shows up

A brief clinical scenario is followed by a question that could have been answered identically without the scenario. The distractors do not require the candidate to use the clinical information to discriminate. The Bloom's level is typically 2 when it could be 3 with modest revision.

Example question IDs

c2b91fa4 — Menkes disease vignette with kinky hair, hypotonia, and seizures. The question asks for the enzyme deficiency, which any candidate who has memorized "Menkes = copper transport defect" can answer without processing the clinical details. The vignette adds no discriminatory function.
d8ef636f — Exercise-induced myoglobinuria vignette. The correct answer is "glycolytic pathway defect" — too vague to require clinical reasoning. The specific diagnosis (McArdle disease / myophosphorylase deficiency) should be the answer, with other glycogenoses as distractors.
40299838 — Neonatal galactosemia with positive Benedict's test. Reasonable framing but the clinical details (jaundice, cataracts, reducing substance in urine) are a textbook triad that maps directly to a memorized label. No integration required.
7bda2d25 — Maternal PKU consequences. The concept is high-yield but the question is presented without any maternal scenario; the clinical framing is entirely absent despite the topic being inherently clinical.
0d90b410 — Von Gierke disease vignette with hepatosplenomegaly, hypoglycemia, hyperuricemia, and lactic acidosis. Reasonable but the answer (glucose-6-phosphatase deficiency) is retrievable by pattern-matching the classic tetrad without reasoning through the biochemical mechanism.
11675199 — MSUD question with the correct concept (α-keto acid decarboxylase deficiency) but no neonatal vignette to anchor the reasoning.
ce30a52b — Refsum's disease and alpha-oxidation. Correct concept, but presented as bare recall rather than a clinical scenario with phytanic acid accumulation and neurological features.

Recommended disposition

Fix. These items have sound factual cores and the topic areas are high-yield. The remediation path is to add a functional clinical stem — one where the clinical details are necessary to select the correct answer rather than merely decorative. The benchmark questions (1ebafee0, c218f1b8, b9c5c07c) provide the template. Priority should be given to items covering lysosomal storage diseases, urea cycle disorders, and aminoacidopathies, as these are the most exam-frequent topics in this category.

5. Fringe, Obsolete, and Misclassified Content

Why this pattern is bad

This category covers three related but operationally distinct problems that share a common consequence: the question occupies a slot in the pool without contributing to exam preparation. Fringe content tests obscure facts with no PYQ precedent and no clinical application (e.g., the SiR stain for tubulin, Bodansky units for ALP, perfluorocarbon oxygen carriers). Obsolete content tests facts that were once exam-relevant but are no longer tested or are factually unstable in current literature (e.g., "most potent antioxidant" as a single-answer question, number of genes in the human genome). Misclassified content is placed under the wrong topic, distorting topic-level analytics and potentially causing the item to be served in the wrong context.

How it shows up

Fringe items typically have no exam tags, no template memberships, and test facts that do not appear in any recent INICET or NEET-PG PYQ. Obsolete items may have older PYQ tags (NEET-PG 2012, 2013) but test facts that have been superseded or are now contested. Misclassified items have a topic label that does not match the content — a mineral cofactor question filed under "Vitamins and Coenzymes," a physics question filed under "Biochemical Techniques," a coagulation question filed under "Hemoglobin and Iron Metabolism."

Example question IDs

8d106fa5 — SiR stain for tubulin. Obscure fluorescence microscopy reagent, no PYQ precedent, no clinical relevance. Options describe histopathology stains from a different discipline.
15fc7c45 — Normal ALP in Bodansky units. Obsolete unit of measurement with no modern clinical application.
b6f8fb6c — Perfluorocarbon as synthetic oxygen carrier. Fringe biochemistry with no exam precedent.
7881d50f — Number of genes in the human genome (~30,000). Factually unstable (ENCODE data revises estimates continuously) and educationally low-value.
3860adbf — "Most potent antioxidant" as a single-answer question. The claim that Vitamin E is the most potent antioxidant is contested in current literature; the question is outdated in framing.
85e1211d — Electromagnetic radiation in spectroscopy, with template membership in "ICSE Physics Electricity Quiz." This is a physics question imported from a non-medical source. Wrong subject entirely.
318a1aaf — Primary characteristic of the Atkins diet. Nutrition trivia with no biochemical mechanism tested and no PG entrance relevance.
4e84a8fb — DNase I produces single-strand nicks. Obscure molecular biology detail with no clinical correlate and no PYQ precedent.
8e9fa786 — Mg²⁺ for kinases filed under "Vitamins and Coenzymes." Correct fact, wrong topic.
8d58ef1c — Vitamin K-dependent clotting factors filed under "Hemoglobin and Iron Metabolism." Correct fact, wrong topic.
2938f797 — Bilirubin estimation wavelength with template membership in "JKBOSE Class 11 Physics/Chemistry." Cross-subject contamination from a school-level science database.

Recommended disposition

Disable fringe and obsolete items outright — they do not serve the exam preparation purpose and some are factually unstable. Misclassified items that are otherwise sound should be fixed by correcting the topic tag; no content change is needed. Items with cross-subject contamination (physics questions, school-level science) should be disabled and flagged for source audit to prevent recurrence.

6. Inborn Errors of Metabolism: Duplicate Concept Pairs and Uneven Depth

Why this pattern is bad

Inborn errors of metabolism (IEM) — lysosomal storage diseases, urea cycle disorders, aminoacidopathies, organic acidemias — are among the highest-frequency topics in INICET and NEET-PG biochemistry. The benchmark set has three IEM questions (CPS-I deficiency, PKU management, Tay-Sachs/hexosaminidase A), and the PYQ set includes Pompe disease and Lesch-Nyhan. The candidate pool has extensive IEM coverage, but it is uneven in a specific way: the same disease is often tested twice in the same shard — once as a bare-recall item and once as a clinical vignette — and the recall version adds nothing when the vignette version exists. Simultaneously, some IEM topics are covered only at recall level with no vignette version available.

How it shows up

Duplicate concept pairs appear where two questions test the same enzyme-disease association at different Bloom's levels. The recall version is redundant when the vignette version exists. Separately, some IEM questions test the enzyme name without testing the accumulated substrate, the inheritance pattern, or the clinical management — the three dimensions that actually appear in exam questions.

Example question IDs

f176ec0f and c77a8ece — Both test alpha-L-iduronidase deficiency in Hurler syndrome. f176ec0f is a bare-recall item; c77a8ece is a well-constructed clinical vignette with organomegaly, corneal clouding, and coronary deposits. The recall version is redundant and should be disabled.
656412aa — "Homogentisate accumulates in alkaptonuria." Single-fact recall. The concept is high-yield but the question adds nothing when a clinical vignette version (dark urine, ochronosis, arthritis) would be far more useful.
b0dc5b8d — Alkaptonuria enzyme (homogentisate oxidase). Bare recall, Bloom's 1. Same concept as above, different angle, same problem.
8c25d144 — Sphingomyelin accumulation in Niemann-Pick disease. Acceptable Bloom's 3 item with appropriate distractors (Tay-Sachs, Gaucher, Fabry). Keep.
c2b91fa4 — Menkes disease vignette that resolves at recall level (see Category 4). The vignette shell is present but the reasoning demand is not.
298aeaeb — Lysosomal storage disorders and mental retardation with "All of the Above" as the correct answer (see Category 3). The concept is valid but the construction is flawed.
3ee2bac8 — Chronic granulomatous disease linked to pentose phosphate pathway defect. Clinical vignette, Bloom's 4, appropriate distractors. This is the quality standard for IEM questions in this pool.

Recommended disposition

Where a duplicate concept pair exists (recall + vignette on the same disease), disable the recall version and keep the vignette version. Where only a recall version exists for a high-yield IEM topic, flag for vignette upgrade rather than disabling — the concept deserves coverage at the appropriate depth. The priority upgrade list should include: alkaptonuria (clinical presentation), Menkes disease (mechanistic reasoning), MSUD (neonatal vignette), and organic acidemias (odor-disease associations upgraded to clinical scenarios). The benchmark questions on CPS-I deficiency and PKU management provide the template.

7. Nutrition and Food Composition Trivia: A Distinct Low-Yield Cluster

Why this pattern is bad

A recurring cluster of questions tests food composition facts — highest calcium food, highest PUFA oil, nitrogen balance in pregnancy, macronutrient content of specific foodstuffs — that have no clinical application in the PG entrance context and no representation in recent INICET or NEET-PG PYQs. These are distinct from clinically relevant nutrition questions (e.g., vitamin deficiency presentations, malnutrition assessment, dietary management of metabolic disorders) and should be treated as a separate category. Some items in this cluster also have factual accuracy concerns: the claim that dates have the highest calcium content is contestable against standard Indian food composition tables (ragi/finger millet has substantially higher calcium per 100g).

How it shows up

Questions ask for the food item with the highest content of a specific nutrient, the macronutrient composition of a named food, or the nitrogen balance status in a specific physiological state — all without any clinical scenario or patient context.

Example question IDs

fbe21a5e — "Food item with highest calcium — Dates." Nutritional trivia, factually suspect (ragi has higher calcium), no clinical application.
5080f3f6 — "Highest PUFA oil." Food composition recall with no clinical relevance.
c39ff67b — Nitrogen balance in pregnancy. Tagged UPSC-CMS, misaligned with INI-CET/NEET-PG standards.
b88493f4 — Macronutrient content of foodstuffs. Rote memorization of food tables.
1a81df96 — Zinc deficiency features. The option construction is weak (one option is a combination of two others) and the question adds no value beyond a list-recall item.

Recommended disposition

Disable the food composition trivia items. They do not appear in the benchmark or PYQ sets, they have no clinical application, and some are factually contestable. Clinically relevant nutrition questions — vitamin deficiency presentations, dietary management of PKU or galactosemia, malnutrition assessment criteria — should be retained and, where possible, upgraded to vignette format. The distinction between "food trivia" and "clinical nutrition" should be applied consistently across the full pool.

Prioritization

The seven issue categories identified above are not equally urgent. The following prioritization reflects both the severity of harm and the operational effort required for remediation.

Tier 1 — Immediate action required (before any further deployment)

Category 2 (Factual Errors and Wrong Answer Keys) is the highest priority. Items with confirmed wrong keys (f073adfe, 762bd09e, 108889f2, b190c2ad, 3ac8afe1, d3cc71cf, be2dd36d) must be disabled immediately. A systematic expert review of the full Biochemistry pool for answer key accuracy is warranted — the error rate in the 200-question sample suggests this is not an isolated problem.

Category 3 (Structural MCQ Construction Failures) is equally urgent for the subset involving missing images (62cd585c, 102648d4) — these questions are literally unanswerable as delivered and must be disabled immediately. Duplicate option errors (9d940927, 340440db) should also be corrected before any further use.

Tier 2 — High priority (next editorial cycle)

Category 1 (Recall-Only Bare-Stem Questions) represents the largest volume of work. The recommended approach is to batch-disable the clearest cases (pure definitional recall with no clinical hook, Bloom's 1, no exam tags) and flag the remainder for vignette upgrade. Given the scale (~60–65% of the sample), this requires a systematic pass rather than item-by-item review.

Category 5 (Fringe, Obsolete, and Misclassified Content) should be addressed in the same editorial cycle. Fringe and obsolete items are straightforward disables. Misclassified items require only a topic tag correction and can be batched efficiently.

Tier 3 — Planned improvement (subsequent editorial cycle)

Category 4 (Thin Clinical Vignettes) and Category 6 (IEM Duplicate Pairs) require more editorial investment — rewriting stems, adding clinical context, resolving duplicate pairs — but the underlying concepts are sound and the items are worth saving. These should be queued for a dedicated upgrade pass with the benchmark questions as the template.

Category 7 (Nutrition Trivia) is a straightforward disable batch with low editorial overhead. It can be processed alongside Category 5.

Example Keep / Fix / Disable Calls

The following calls are drawn directly from the reviewed sample and are intended to illustrate the application of the issue categories above.

KEEP

Question ID	Topic	Rationale
`c77a8ece`	Genetic Disorders	Well-constructed Hurler syndrome vignette with multi-system clinical scenario, Bloom's 3, appropriate distractors. Meets benchmark standard.
`3ee2bac8`	Carbohydrate Metabolism	Chronic granulomatous disease linked to pentose phosphate pathway defect. Clinical vignette, Bloom's 4, discriminating distractors.
`de2b61c2`	Carbohydrate Metabolism	Pyruvate carboxylase in gluconeogenesis during prolonged fast. Applied Bloom's 4, physiological context, plausible distractors.
`fcbc861b`	Amino Acid Metabolism	Urea cycle defect vignette with elevated argininosuccinate pointing to argininosuccinate lyase deficiency. Bloom's 3, mirrors benchmark 1ebafee0.
`c02bbafb`	Vitamins and Coenzymes	Biotin/pyruvate carboxylase/hypoglycemia vignette. Bloom's 3, mechanistic reasoning required, PYQ-tagged.
`f71144fa`	Lipid Metabolism	HMG-CoA reductase regulation EXCEPT question. Tests SREBP/SRE/Insig1/phosphorylation state. Genuine discrimination at Bloom's 2–3.
`01d47f7a`	Carbohydrate Metabolism	G6PD/pentose phosphate pathway — mechanistic stem linking enzyme deficiency to NADPH loss to hemolytic anemia. Bloom's 3.
`ddf7326a`	Metabolic Integration	Pyruvate substrates — requires knowledge of transamination specificity. Bloom's 4 rating justified.
`8c25d144`	Genetic Disorders	Sphingomyelin accumulation in Niemann-Pick. Bloom's 3, classic differential distractors (Tay-Sachs, Gaucher, Fabry).
`fd25b3f6`	Amino Acid Metabolism	Sweaty feet odor/isovaleric acidemia. Classic PYQ-style association, well-formed distractors covering major aminoacidopathies.
`c451455d`	Enzymes	Glucokinase properties (high Km, not inhibited by G6P). Tests nuanced enzyme kinetics concept, PYQ-adjacent.
`13b2f065`	Molecular Biology	Dihydropyrimidine dehydrogenase deficiency and 5-fluorouracil toxicity. Pharmacogenomics-biochemistry interface, high-yield applied concept.

FIX

Question ID	Issue	Required Action
`c76cd96d`	Answer key ambiguity — option A ("occurs primarily in the liver") is also correct	Rewrite to make only one option unambiguously correct; verify against primary source. Priority fix.
`d3cc71cf`	Wrong answer key — XP attributed to DNA Pol I/ligase instead of NER endonucleases	Correct answer key to NER components; rewrite options to reflect XPC, ERCC2, endonuclease incision.
`9d940927`	Duplicate option — "1.5 ATP" appears twice	Replace one duplicate with a distinct distractor (e.g., "3.5 ATP").
`340440db`	Duplicate option + wrong stem category — folic acid appears twice; stem asks for "amino acid" but answer is a vitamin	Remove duplicate option; rewrite stem to ask for "coenzyme/carrier of one-carbon units."
`ffbaaeab`	Thiamine deficiency — "memory loss" (Korsakoff) listed as distractor but is also a valid thiamine deficiency manifestation	Replace "memory loss" with a non-thiamine distractor (e.g., night blindness) or restrict stem to a specific clinical scenario.
`34c06835`	Near-synonym distractor — "Termination mutation" is near-synonym of correct answer "Nonsense mutation"	Replace "Termination mutation" with "Frameshift mutation."
`b1066e9f`	Near-duplicate distractor pair — options B and C differ only by stimulated/inhibited for CAT-I	Rewrite one option to test a different aspect of beta-oxidation regulation.
`d8ef636f`	Thin vignette — correct answer "glycolytic pathway defect" is too vague	Replace with specific enzyme-level answer (McArdle disease / myophosphorylase deficiency) and use other glycogenoses as distractors.
`c2b91fa4`	Thin vignette — Menkes disease resolves at recall level	Add mechanistic question stem: "Which enzyme's activity is most directly impaired?" to push to Bloom's 3.
`efe180c8`	Convoluted negative framing + contestable correct answer for ribozymes	Rewrite as direct positive stem: "Which of the following is a function of ribozymes?" Replace misleading correct option with peptidyl transferase activity.
`0deb82fc`	Multiple simultaneously correct options — progesterone and estrogen are also non-cAMP hormones	Restrict options to hormones that clearly discriminate cAMP vs. non-cAMP signaling; rewrite stem to specify "which acts via JAK-STAT."
`8d58ef1c`	Misclassified topic — Vitamin K-dependent factors filed under "Hemoglobin and Iron Metabolism"	Reclassify to "Vitamins and Coenzymes." No content change needed.
`2938f797`	Cross-subject contamination — template membership includes JKBOSE Class 11 Physics/Chemistry	Remove non-medical template memberships; verify clinical context (specify Jendrassik-Grof vs. direct absorbance method in stem).
`13fcfe97`	Factually invalid distractor — "Factor VI" does not exist as a recognized coagulation factor	Replace "Factor VI" with Factor V or Factor XIII; reclassify topic from "Nutrition and Metabolism" to "Vitamins and Coenzymes."

DISABLE

Question ID	Issue	Rationale
`f073adfe`	Wrong answer key — net ATP from glycolysis listed as 7	Straightforward factual error; actively misinforms learners. Disable until corrected.
`762bd09e`	Wrong answer key — ascorbic acid marked as decreasing iron absorption	Factual inversion; ascorbic acid enhances iron absorption.
`108889f2`	Wrong answer key — albumin marked as mitochondrial fatty acid transport mechanism	Albumin is a plasma carrier; carnitine shuttle is the correct answer.
`3ac8afe1`	Wrong stem — protein marked as the only macromolecule requiring a template	DNA and RNA also require templates; question is factually incorrect as framed.
`866a9a94`	Factually contestable — immune complex formation attributed to quaternary structure	Antigen-antibody binding involves tertiary surface complementarity; risks teaching an error.
`62cd585c`	Missing image + "All of the Above" key	Unanswerable without the pedigree image; additionally uses a disqualifying answer format.
`85e1211d`	Wrong subject — physics question filed under Biochemical Techniques	Template membership in "ICSE Physics Electricity Quiz" confirms cross-subject contamination.
`f176ec0f`	Duplicate concept — bare-recall Hurler syndrome item	Redundant with vignette version `c77a8ece`; recall version adds nothing.
`8d106fa5`	Fringe content — SiR stain for tubulin	Obscure research-lab technique, no PYQ precedent, no clinical relevance. Options describe histopathology stains from a different discipline.
`15fc7c45`	Obsolete content — ALP in Bodansky units	Obsolete unit of measurement with no modern clinical application.
`7881d50f`	Factually unstable + low-yield — number of genes in human genome	Estimate is continuously revised; no clinical relevance.
`fbe21a5e`	Factually suspect + low-yield — dates as highest calcium food	Ragi has substantially higher calcium per 100g by standard Indian food composition tables; no clinical application.
`97c8426a`	Pure recall — transcription definition	Zero discriminatory value at PG level; no candidate who has reached PG preparation should miss this.
`cd56e8d9`	Pure recall — ceruloplasmin metal	Bloom's 1, no clinical hook, no discriminatory value.
`6eb16a69`	Pure recall — Fe²⁺ atoms in haemoglobin	Bloom's 1, trivially easy, no reasoning required.
`20fe8c9a`	Chemistry arithmetic — molecular weight difference between phenylalanine and tyrosine	Not a biochemical reasoning item; no clinical or conceptual value for PG medical entrance.
`a80ec08d`	Pure recall — karyotyping detects chromosomal abnormalities	Self-evident; zero discriminatory value.
`9dcc072c`	Pure recall — tertiary protein structure definition	Verbatim textbook definition restated as a question.
`44c605a3`	Pure recall — SDS-PAGE separates by size	Trivial single-fact recall; redundant with better-framed technique questions in the PYQ set.
`3c8b2a11`	Incoherent distractor set + factual error in stem	Three of four options are amino acids in a question about chaperone function; stem contains factual inaccuracy about "reverse folding."
`8cb18f34`	Unfalsifiable correct answer — second messengers defined as "substances that increase or decrease cellular function"	So broad it describes almost any molecule; tests nothing meaningful.
`318a1aaf`	Fringe content — Atkins diet primary characteristic	Nutrition trivia with no biochemical mechanism tested and no PG entrance relevance.
`4e84a8fb`	Fringe content — DNase I single-strand nicks	Obscure molecular biology detail, no clinical correlate, no PYQ precedent.
`3b45265e`	Fringe content — Xanthoproteic test	Narrow colorimetric test trivia, Bloom's 1, no clinical application, no PYQ precedent at INI-CET/NEET-PG level.