Biochemistry Question Quality Review

Executive Summary

This review covers a candidate sample of 100 validated non-gold Biochemistry questions (60 generic + 40 risky), evaluated against 8 benchmark items and 12 recent PYQs. The subject pool contains 10,646 questions total.

The most striking finding in this sample is the severe Bloom's compression at the bottom of the taxonomy. Of the 100 candidate questions, 55 are coded Bloom's Level 1 and 44 are Level 2, with zero items at Levels 3 or 4 and only one at Level 5. The benchmark set, by contrast, operates entirely at Bloom's Level 3 with rich clinical vignettes. This is not a marginal gap — it represents a structural mismatch between what the content bank is producing and what INICET/NEET-PG examinations actually test.

Beyond the Bloom's problem, the reviewed set contains multiple factually unsafe items (wrong keys, misleading distractors), a cluster of broken or malformed stems, and a large volume of low-yield trivia that would not survive contact with a modern PG exam. There are also subject contamination issues, with questions that belong more naturally in Physiology, Pharmacology, or Pathology appearing under Biochemistry topic tags.

Summary counts across the 100-question candidate sample:

Issue Category	Approximate Count	Recommended Action
Wrong Key or Factually Unsafe	8–10	Disable or urgent fix
Wrong Subject / Topic Placement	4–5	Reassign or disable
Broken Delivery	2–3	Fix or disable
Low-Value But Correct	40–45	Disable (gold coverage exists)
Repetitive / Duplicative Coverage	6–8	Disable duplicates
Worthwhile Concept, Weak Execution	12–15	Fix stem/options

The overall usable yield from this 100-question sample, without remediation, is estimated at roughly 25–30 questions. The rest require either immediate disabling or substantive rewriting before they can be deployed in any test template.

What Good Looks Like

The benchmark items establish a clear quality bar that the candidate sample largely fails to meet. The defining features of a high-quality Biochemistry item for INICET/NEET-PG are:

1. Clinical vignette as the entry point. Every benchmark question opens with a patient scenario — age, sex, presenting complaint, examination findings, and relevant investigations. The biochemical concept is reached through clinical reasoning, not direct recall. For example, benchmark item f4261145 presents a 45-year-old alcoholic with confusion, ataxia, and ophthalmoplegia before asking about thiamine. The candidate must recognize Wernicke's encephalopathy and understand why dextrose without thiamine is dangerous. The concept tested (thiamine as cofactor for pyruvate dehydrogenase) is identical to what a bare-stem question would test, but the vignette forces application rather than recognition.

2. Bloom's Level 3 minimum. All eight benchmark items are coded at Bloom's Level 3 (application) or higher. The PYQ set includes items at Levels 3 and 4. The candidate sample has no items at Level 3 or above among the 100 reviewed questions. This is the single most important gap.

3. Distractors that represent plausible errors. In benchmark item 1ebafee0, the four options are all urea cycle enzymes — CPS-I, OTC, argininosuccinate synthetase, and argininosuccinate lyase. Distinguishing between them requires knowing the specific amino acid profile (low citrulline and arginine pointing to a proximal defect). This is discriminating. In contrast, the candidate sample frequently uses distractors that are obviously wrong to any student who has opened a textbook (e.g., "Phosphofructokinase" as a distractor for the rate-limiting enzyme of cholesterol synthesis in cdba3715).

4. Single unambiguous correct answer with a defensible biochemical rationale. The benchmark items have keys that are not merely "most correct" but are mechanistically precise. The candidate sample contains several items where the key is contestable or where a distractor is partially correct.

5. Exam-relevant topics. Benchmark and PYQ items cluster around metabolic integration, inborn errors of metabolism, vitamins as coenzymes, molecular biology applied to disease, and clinical biochemistry markers. The candidate sample devotes substantial space to enzyme classification trivia, food composition facts, and basic definitions that have not appeared in INICET or NEET-PG in recent years.

Main Issue Categories

1. Wrong Key or Factually Unsafe

Why this pattern is bad

A wrong key is the most serious quality defect in any MCQ bank. It actively harms candidates who study from the platform, reinforces incorrect knowledge, and creates medico-legal exposure if the question appears in a scored test. Items in this category must be treated as urgent, not deferred.

How it shows up

In this sample, wrong-key problems appear in two forms: (a) the marked correct answer is factually incorrect by standard biochemistry references, and (b) the marked correct answer is defensible but a distractor is equally or more correct, making the item ambiguous rather than wrong in the strict sense. Both are unsafe for deployment.

Example question IDs and explanations

16b8f824 — MELAS and ETC Complex The question asks which ETC complex is inhibited in MELAS syndrome. The marked key is Complex II. This is factually incorrect. MELAS (Mitochondrial Encephalopathy, Lactic Acidosis, and Stroke-like episodes) is caused by mutations in mitochondrial DNA, most commonly the m.3243A>G mutation in the MT-TL1 gene encoding tRNA-Leu, which impairs the assembly and function of Complex I (NADH dehydrogenase). Complex II is encoded entirely by nuclear DNA and is not implicated in MELAS. This is a clear wrong key. Any candidate who knows the correct answer (Complex I) will be penalized. Disposition: Disable immediately.

ef7b1ad7 — ATP III Classification, Total Cholesterol The question asks for the desired level of total cholesterol per ATP III. The marked key is "< 40 mg/dL." This is the ATP III threshold for low HDL-cholesterol in men, not total cholesterol. The desired total cholesterol per ATP III is < 200 mg/dL, which appears as a distractor. This is a straightforward key-distractor swap — the correct answer is in the option list but is not marked correct. Disposition: Disable immediately (or fix key if the stem is otherwise retained, but the stem itself is low-yield — prefer disable).

756c4d7e — HbA1c Timeframe The marked key is "6–8 weeks." Standard references (Harper's, Lehninger, ADA guidelines) consistently state that HbA1c reflects glycemic control over the preceding 2–3 months (approximately 8–12 weeks), with the most commonly cited figure being 8–12 weeks or "the preceding 3 months." The option "6–8 weeks" is a truncated and potentially misleading figure. The option "14–18 weeks" is also present and is closer to the full RBC lifespan. This item is borderline unsafe — the key is not definitively wrong but is the least precise of the available options and will confuse candidates who have learned the standard 3-month figure. Disposition: Fix — rewrite options to include "8–12 weeks" or "2–3 months" as the correct answer, or disable if the concept is already well-covered in gold items.

a4b6d7db — Thiamine Pyrophosphate and Transamination The question asks which enzymatic reaction requires TPP (thiamine pyrophosphate) as a coenzyme. The marked key is "Transamination." This is factually incorrect. Transamination reactions require pyridoxal phosphate (PLP, vitamin B6), not TPP. TPP is the coenzyme for oxidative decarboxylation reactions: pyruvate dehydrogenase, α-ketoglutarate dehydrogenase, branched-chain α-keto acid dehydrogenase, and transketolase. All three of the other options listed (branched-chain amino acid dehydrogenase complex, 2-hydroxy phytanoyl-CoA lyase, transketolase) are legitimate TPP-dependent reactions. The correct answer is present in the distractors but not marked. This is a dangerous wrong key because it directly contradicts a high-yield PYQ concept (pyridoxine required for transamination, tested in d2153405). Disposition: Disable immediately.

ad1b1d0b — Causes of Metabolic Alkalosis The question asks which is NOT a cause of metabolic alkalosis. The marked key is "Both vomiting and renal failure." The logic appears to be that renal failure causes metabolic acidosis (not alkalosis), making it the "not a cause" answer, and the question is trying to use a combined option. However, the option structure is internally contradictory: if renal failure is not a cause of metabolic alkalosis, then "both vomiting and renal failure" cannot be the answer to "which is NOT a cause" — because vomiting IS a cause. The question is asking for what is NOT a cause, and the answer "both vomiting and renal failure" implies neither is a cause, which is wrong for vomiting. The stem and option logic are broken. Additionally, "fever" as an option is a distractor that does not cleanly fit either category. This item is ambiguous and potentially misleading. Disposition: Disable — the option logic is broken and cannot be fixed without a full rewrite.

209f3c5d — Amino Acid Deficiency in Pulses The marked key is "Cysteine." Standard nutritional biochemistry references (Harper's, Devlin) consistently state that pulses (legumes) are deficient in methionine (and by extension cysteine, since cysteine is synthesized from methionine), but the primary limiting amino acid in pulses is methionine, not cysteine. Methionine is listed as a distractor and is not marked correct. This is a factual inaccuracy at the level of the key. Disposition: Fix key to methionine, or disable. Given that this is a low-yield nutrition trivia item, prefer disable.

324832c9 — Ammonia and Plaque Formation The question states that ammonia causes "decrease in plaque formation" and marks this as correct. This is a dental biochemistry claim that is not standard content in any major Indian PG Biochemistry textbook. The relationship between ammonia (produced by urease-positive oral bacteria) and dental plaque/calculus is that ammonia raises oral pH, which promotes calculus (tartar) formation, not plaque reduction. The marked key appears to be incorrect or at minimum is drawn from a non-standard source. This item also represents subject contamination (dental biochemistry). Disposition: Disable.

883c2bea — Micronutrient Deficiency Causing Anemia The marked key is "Copper." While copper deficiency can cause anemia (hypochromic, microcytic, resembling iron deficiency), the question is dangerously non-specific. Iron deficiency, B12 deficiency, folate deficiency, and B6 deficiency all cause anemia and are far more commonly tested. The question has no stem context to narrow the answer, and the distractors (molybdenum, selenium, fluorine) are chosen to make copper look correct by elimination rather than by knowledge. The item tests trivia rather than understanding and the key, while not strictly wrong, is misleading in isolation. Disposition: Disable — low-yield, non-specific, and the distractor set is designed to trick rather than discriminate.

2. Wrong Subject or Wrong Topic Placement

Why this pattern is bad

Subject contamination inflates apparent Biochemistry coverage while actually testing content that belongs in Physiology, Pharmacology, Pathology, or Dentistry. It misleads candidates about what to expect in Biochemistry-tagged tests and creates redundancy with other subject banks. It also distorts topic-level analytics.

How it shows up

In this sample, the contamination is observed in two forms: (a) questions that are unambiguously from another subject and have been tagged to Biochemistry, and (b) questions that sit at the border of Biochemistry and another subject but are framed in a way that makes them more appropriate for the other subject.

Example question IDs and explanations

9a2f471d — Becker Muscular Dystrophy Inheritance The question asks for the mode of inheritance of Becker muscular dystrophy. While dystrophin is a protein and the gene defect is biochemically relevant, the question as written is a pure genetics/Pathology question about inheritance pattern. It does not engage any biochemical concept — no enzyme, no metabolic pathway, no molecular mechanism. This belongs in Pathology or Genetics, not Biochemistry. Compare with benchmark item b9c5c07c (HGPRT deficiency, Lesch-Nyhan), which asks about inheritance but is anchored in a biochemical enzyme deficiency with a clinical vignette. 9a2f471d has no such anchor. Disposition: Reassign to Genetics/Pathology, or disable from Biochemistry bank.

324832c9 — Ammonia and Dental Plaque As noted above, this is dental biochemistry content that does not appear in standard Indian PG Biochemistry curricula. It belongs, if anywhere, in Dental Biochemistry or Oral Biology. Disposition: Disable from Biochemistry bank.

f71c0096 — Dietary Fiber in Type 2 Diabetes The question asks what dietary component is crucial for managing type 2 diabetes. The answer is dietary fiber. This is a Nutrition/Medicine/Physiology question. It does not test any biochemical mechanism. The stem is vague ("what type of dietary component is crucial") and the answer is a clinical nutrition recommendation, not a biochemical concept. Disposition: Disable from Biochemistry bank.

d685f75b — POMC Derivative The question asks which hormone is derived from POMC. While POMC processing is a legitimate biochemistry topic (post-translational processing, signal peptide cleavage), the question is framed as a pure recall item about hormone identity with no biochemical context. It sits more naturally in Physiology (endocrinology) or Pharmacology. The distractors (acetylcholine, dopamine, norepinephrine) are not POMC derivatives and are trivially eliminated, making this a low-discrimination item regardless of subject placement. Disposition: Disable — low-yield, wrong subject framing, trivial distractors.

3. Broken Delivery (Missing Image, Malformed Options, Incomplete Stem)

Why this pattern is bad

Broken delivery makes a question unscorable or misleading regardless of whether the underlying concept is valid. Malformed option logic is particularly dangerous because it can make a wrong answer appear correct through process of elimination.

How it shows up

In this sample, broken delivery appears primarily as malformed option structures (combined options that create logical contradictions) and incomplete or ambiguous stems.

Example question IDs and explanations

3614e2a9 — Collagen Type and Anchoring Fibrils The stem reads: "Collagen ------ forms the main part of anchoring fibrils in epithelial tissues?" The blank in the stem is a formatting artifact — this appears to be a fill-in-the-blank question that was converted to MCQ format without cleaning the stem. The double question mark at the end is also a formatting error. The stem is technically parseable but looks unprofessional and may confuse candidates. The underlying concept (Collagen VII in anchoring fibrils) is correct and testable. Disposition: Fix stem — remove the blank placeholder and rewrite as a clean MCQ stem. The concept is worth keeping.

ad1b1d0b — Metabolic Alkalosis (also cited under Wrong Key) As described above, the option "Both vomiting and renal failure" as the answer to "which is NOT a cause" creates a logical contradiction. This is a broken option structure, not merely a wrong key. Disposition: Disable.

b223b255 — Zinc Deficiency The option structure places "All of the below" as option A (the first option listed), with the individual features as options B, C, and D. This is a non-standard "all of the above" construction where the umbrella option appears first rather than last. While technically answerable, this format is confusing and non-standard for Indian PG MCQs. It also means that a candidate who reads only option A and recognizes it as likely correct never needs to evaluate the individual options, reducing the item's discriminating power. Disposition: Fix — restructure so individual options appear first and "All of the above" appears last, or rewrite as a standard positive stem.

c57ef6ee — Transgenic Mice The marked key is "All of the above." All-of-the-above keys are generally considered poor MCQ practice because they reward partial knowledge (a candidate who knows any one of the three individual statements is true can select "all of the above" without knowing the others). This is a structural quality issue that appears in multiple items in the sample (1b37da80, b22bde12, c57ef6ee). The transgenic mice item is additionally low-yield for INICET/NEET-PG. Disposition: Disable — low-yield content and broken option structure.

4. Low-Value But Correct (Too Simple, Low-Yield, Trivia-Heavy, Weak Exam Relevance)

Why this pattern is bad

This is the largest single category in the reviewed sample. Questions that are factually correct but test only rote recall of isolated facts at Bloom's Level 1 do not prepare candidates for INICET or NEET-PG, which have moved decisively toward application and clinical reasoning. Deploying these items in test templates creates a false sense of preparation, depresses the average difficulty of practice tests, and wastes slots that could be occupied by high-quality items. When strong gold-standard coverage already exists for the same concept, there is no justification for retaining a weaker version.

How it shows up

This pattern dominates the candidate sample. Approximately 40–45 of the 100 reviewed questions fall into this category. They share the following features: no clinical context, single-fact stem, distractors that are obviously wrong to any prepared candidate, and Bloom's Level 1 coding. Many test concepts that are covered far better by existing PYQs or benchmark items.

Example question IDs and explanations

cdba3715 — Rate-Limiting Enzyme of Cholesterol Synthesis "In cholesterol synthesis, which enzyme is rate-limiting?" with options HMG-CoA reductase, HMG-CoA synthetase, 7-alpha hydroxylase, and phosphofructokinase. This is a single-fact recall item. The answer (HMG-CoA reductase) is the first thing any student learns about cholesterol synthesis. The distractor "phosphofructokinase" is from a completely different pathway and is trivially eliminated. The PYQ set already contains a better item on familial hypercholesterolemia (827a667c) that tests the same pathway at a higher level. No clinical context, no reasoning required. Disposition: Disable — concept covered by better gold items.

e23e926f — Vitamin K and Thrombin "Which vitamin is required for the production of thrombin?" This is a Bloom's Level 1 recall item. The answer (Vitamin K) is universally known. The risky sample contains a better item on the same concept (ccbdccc3) that specifies gamma-carboxylation of glutamic acid in clotting factors, which is at least one level more specific. Neither item reaches the quality bar of the benchmark set, but e23e926f is the weaker of the two. Disposition: Disable — concept covered by slightly better item in the same sample, and neither reaches benchmark quality.

e5b8ae37 — Anticodon Region "The anticodon region is an important part of the: r-RNA / mRNA / t-RNA / hn-RNA." This is a first-year MBBS question. It has no place in a PG entrance preparation bank. The answer is universally known to any medical student. Disposition: Disable.

de76256c — 21st Amino Acid "Which is the 21st amino acid?" with answer Selenocysteine. This is pure trivia. Selenocysteine is a legitimate biochemistry topic (it is incorporated co-translationally via a UGA codon recoded by SECIS elements), but the question tests only the number "21" rather than any mechanistic understanding. It has not appeared in recent INICET or NEET-PG papers in this form. Disposition: Disable — trivia with no clinical or mechanistic relevance.

77b14ab5 — Hydrophobic Amino Acid "Which of the following is a hydrophobic amino acid? Alanine / Tyrosine / Glycine / Histidine." Bloom's Level 1, no clinical context, trivially answered by elimination (tyrosine has a hydroxyl group, histidine has an imidazole ring, glycine is the simplest amino acid). This tests memorization of a classification, not understanding. Disposition: Disable.

ba929cc0 — Vitamin C and Iron Absorption "Which vitamin enhances iron absorption?" This is a standard first-year fact. The answer (Vitamin C) is universally known. No clinical context, no reasoning required. Disposition: Disable.

8b7f5f78 — RFLP and Southern Blotting "Restriction fragment length polymorphism uses which of the following techniques? Southern blotting / Northern blotting / Western blotting / Eastern blotting." This is a Bloom's Level 1 recall item. The answer is correct (Southern blotting), but the question tests only a name association. The PYQ set contains a better molecular biology item (7d20c9a3, PCR cycles calculation) that tests actual quantitative reasoning. The blotting technique associations are better tested in a clinical context (e.g., "HIV diagnosis uses which blotting technique?"). Disposition: Disable — concept can be tested at higher Bloom's level with a clinical anchor.

6aa10546 — Trace Elements "All the following are trace elements required by the body except? Zinc / Copper / Selenium / Boron." This is a nutrition trivia item. Boron's status as a required trace element in humans is actually debated in the literature — it is considered essential in plants and some animals, and there is emerging evidence for a role in humans, but it is not universally classified as a required human trace element. The question may be correct by the standard used in Indian textbooks, but it is low-yield and the concept has no clinical application in PG entrance exams. Disposition: Disable.

cb71eabf — Most Common Urea Cycle Enzyme Deficiency "Most common enzyme deficiency in the urea cycle is: OTC deficiency." This is a single-fact recall item. The answer is correct (OTC deficiency is X-linked and the most common urea cycle defect). However, the benchmark set already contains a far superior item on urea cycle defects (1ebafee0) that presents a neonatal case with amino acid profiles and asks for enzyme identification through clinical reasoning. The bare-stem version adds nothing. Disposition: Disable — concept covered by benchmark item 1ebafee0 at a much higher quality level.

a46ba1ef — Water-Soluble Vitamin "Which of the following is a water-soluble vitamin? Vitamin A / Vitamin B12 / Vitamin D / Vitamin E." This is a pre-clinical year question. Vitamins A, D, and E are fat-soluble; this is taught in the first week of biochemistry. No PG entrance exam has asked this in recent years. Disposition: Disable.

4c48fa72 — Lambda Phage Symbol "In molecular biology, λ (lambda) phage is commonly used as a cloning vector. Which symbol correctly represents lambda?" with options being Greek letters. This question tests whether a candidate can identify the Greek letter lambda. It has zero biochemical content. It is the weakest item in the entire reviewed sample. Disposition: Disable immediately.

172fac46 — Immunoglobulin Heavy Chain Gene Locus "Which of the following is true about the location of the immunoglobulin (Ig) heavy chain gene locus? Chromosome 14." This is a chromosomal location fact with no clinical or biochemical reasoning required. It belongs more naturally in Immunology or Genetics. The distractor "All of the above" is also logically impossible since the three chromosomes listed are mutually exclusive locations. Disposition: Disable — wrong subject framing and trivial recall.

e0e63051 — Sorghum and Leucine "Sorghum contains an excess of which amino acid?" This is food composition trivia. The answer (leucine) is relevant because excess leucine in sorghum is proposed to contribute to pellagra in populations dependent on sorghum (by interfering with tryptophan-to-niacin conversion). However, the question does not test this mechanism — it tests only the food-amino acid association. A better question would present a patient with pellagra-like symptoms in a sorghum-dependent population and ask about the biochemical mechanism. Disposition: Disable — trivia form; concept could be salvaged with a clinical vignette rewrite, but the concept is low-priority.

5. Repetitive or Duplicative Coverage

Why this pattern is bad

Duplicate or near-duplicate items within the same bank waste test template slots, create an illusion of coverage breadth, and can cause candidates to see the same question twice in different tests — which degrades the perceived quality of the platform. In a bank of 10,646 questions, duplication is expected, but it should be actively managed.

How it shows up

In this sample, duplication appears both as exact concept repetition (same fact, same options, different question IDs) and as near-duplicate coverage where two items test the same narrow concept at the same Bloom's level with no meaningful differentiation.

Example question IDs and explanations

a428bffd appears in both the generic and risky samples Question ID a428bffd-1bd6-4991-a36e-797ad8818da4 ("Protein disulfide isomerase is involved in?") appears identically in both the generic validated set and the risky validated set. This is a literal duplicate — same question ID, same text, same options, same key. This should not be possible if deduplication is functioning correctly. The item itself is low-yield (Bloom's Level 2, easy), but the duplication is a data integrity issue independent of quality. Disposition: Remove one instance; evaluate the remaining item for quality (likely disable given low yield).

e23e926f and ccbdccc3 — Vitamin K and Clotting e23e926f asks "Which vitamin is required for the production of thrombin?" and ccbdccc3 asks "Gamma-carboxylation of glutamic acid in clotting factors II, VII, and protein C is dependent on which vitamin?" Both have Vitamin K as the correct answer. The second item is marginally better (it specifies the biochemical mechanism), but neither reaches benchmark quality. Together they represent duplicative coverage of the same narrow concept. Disposition: Disable e23e926f; evaluate ccbdccc3 for possible upgrade to a clinical vignette format.

d2153405 and a4b6d7db — Pyridoxine/TPP and Transamination d2153405 correctly states that pyridoxine is required for transamination. a4b6d7db incorrectly marks transamination as a TPP-dependent reaction. These two items are in direct factual conflict with each other. The first is correct; the second has a wrong key. Having both in the bank is dangerous. Disposition: Keep d2153405 (correct, though low-yield); disable a4b6d7db (wrong key).

50e5bffd and 0dffd785 — MPS Enzyme Deficiencies 50e5bffd asks about Hunter syndrome enzyme deficiency (iduronate-2-sulfatase). 0dffd785 asks about Sly syndrome substrate accumulation. Both are Bloom's Level 1 MPS trivia items with no clinical context. The MPS disorders are high-yield for PG exams, but the correct way to test them is through a clinical vignette (as in benchmark item f6a8da17, which presents a child with coarse facies and developmental delay and asks about mannose-6-phosphate tagging). Bare-stem enzyme-name recall items for individual MPS subtypes are low-yield in isolation. Disposition: Disable both — concept better covered by clinical vignette format.

cb71eabf and 2d2b8cc6 — Urea Cycle Defects cb71eabf asks for the most common urea cycle enzyme deficiency (OTC). 2d2b8cc6 asks which urea cycle enzyme deficiency does NOT cause hyperammonemia (ornithine aminotransferase). These are related but distinct concepts. The second item (2d2b8cc6) is the better of the two — it requires understanding of which enzymes are actually in the urea cycle versus adjacent pathways. However, both are still below benchmark quality. Disposition: Disable cb71eabf; consider upgrading 2d2b8cc6 with a clinical vignette.

6. Worthwhile Concept, Weak Execution (Keep the Concept, Fix the Stem/Options/Vignette)

Why this pattern is bad

These items test concepts that are genuinely high-yield for INICET/NEET-PG but are written in a way that reduces their discriminating power, introduces ambiguity, or fails to reach the Bloom's Level 3 standard. Discarding these concepts entirely would be wasteful; the right action is targeted rewriting.

How it shows up

Weak execution appears as: bare-stem recall questions on high-yield topics that should have clinical vignettes; questions with partially correct distractors that create ambiguity; questions where the stem gives away the answer; and questions where the correct answer is correct but the explanation embedded in the options is misleading.

Example question IDs and explanations

8f04fb38 — Glucokinase and Post-Meal Glucose Uptake The concept tested — that glucokinase (hexokinase IV) has a high Km for glucose and is not saturated at normal blood glucose levels, so it responds proportionally to post-meal glucose surges — is genuinely high-yield and has appeared in PYQs. The correct answer ("There is an increase in the phosphorylation of glucose by glucokinase") is defensible. However, the distractor "Glucokinase has a low Km for glucose" is the exact opposite of the truth and is too obviously wrong. A better distractor would be "Glucokinase is induced by insulin" (which is true but is not the primary mechanism for the immediate post-meal response). The stem could also be strengthened by specifying that GLUT-2 is already present at the basolateral membrane and asking why glucose uptake increases despite GLUT-2 being constitutively expressed. Disposition: Fix — rewrite distractors to include plausible errors; add brief clinical context (e.g., "In a patient who has just consumed a high-carbohydrate meal, portal glucose rises to 15 mmol/L...").

52a06b5a — Metabolic Processes Exclusive to Mitochondria The concept (ketone body synthesis occurs exclusively in mitochondria) is correct and testable. However, the question is weakened by the fact that "fatty acid synthesis" and "cholesterol synthesis" are listed as distractors — both occur primarily in the cytosol, which is correct, but "gluconeogenesis" is listed as a distractor implying it is not exclusively mitochondrial, which is also correct (gluconeogenesis spans both compartments). The question is technically answerable but the distractor set does not clearly test understanding of compartmentalization. A better version would ask about a specific step of ketogenesis (e.g., HMG-CoA synthase in the mitochondria versus cytosol) or present a clinical scenario of a fasting patient. Disposition: Fix — rewrite with a clinical vignette (fasting state, elevated ketones) and sharpen the distractor set to test compartmentalization knowledge specifically.

ac8aecda — Porphyria Clinical Presentation The concept is excellent — porphyria presenting with the triad of neuropsychiatric symptoms, abdominal pain, and neuropathy is a classic high-yield clinical biochemistry scenario. The question correctly identifies porphyria as the answer. However, the stem is too easy: the three features listed (axonal neuropathy, unexplained abdominal pain, psychiatric illness) are the textbook triad for acute intermittent porphyria, and the distractors (PNH, arsenic toxicity, hypothyroidism) are not serious competitors. A stronger version would add urine color change, precipitating factors (drugs, fasting), or laboratory findings (elevated ALA and PBG) and ask for the specific enzyme deficiency or the specific type of porphyria. Disposition: Fix — add clinical detail and sharpen to ask for enzyme deficiency (PBG deaminase/hydroxymethylbilane synthase) or specific porphyria type.

db674f87 — Hawkinsinuria and Swimming Pool Odor The concept (Hawkinsinuria, a rare tyrosine metabolism disorder causing a swimming pool-like odor due to accumulation of hawkinsin, a cyclic amino acid) is genuinely high-yield for INICET as an unusual inborn error of metabolism. The question is correctly keyed. However, the stem is a bare recall item ("What inborn error of metabolism is characterized by a swimming pool-like odor?") with no clinical context. A stronger version would present a neonate with metabolic acidosis, failure to thrive, and a characteristic urine odor, and ask for the metabolic pathway involved. Disposition: Fix — add a brief neonatal vignette with urine odor description and ask for the metabolic pathway or enzyme involved.

13655811 — Pyrimidine Metabolism Features The concept is high-yield (pyrimidine synthesis, CAD polypeptide, orotic aciduria, mitochondrial dihydroorotate dehydrogenase, Reye syndrome). The correct answer — that in Reye syndrome there is decreased cytosolic carbamoyl phosphate — is a nuanced point: in Reye syndrome, mitochondrial dysfunction leads to accumulation of carbamoyl phosphate in the mitochondria, which spills over into the cytosol and drives pyrimidine synthesis, resulting in orotic aciduria. The question correctly identifies this as the "NOT a feature" answer. However, the stem is a bare "which is NOT a feature" format with no clinical context. A Reye syndrome vignette (child with viral illness, aspirin use, encephalopathy, orotic aciduria) would make this a Bloom's Level 3 item. Disposition: Fix — add a clinical vignette for Reye syndrome and reframe as an application question.

2943bcd4 — Allosteric vs. Active Site Inhibition The concept (allosteric site targeting for specificity in conserved enzymes) is legitimate and the question is coded at Bloom's Level 5, which is unusual in this sample. However, the execution is poor. The correct answer ("Targeting allosteric sites for better specificity") is stated as a general principle, but the distractor "Effectiveness depends on the specific enzyme and context" is also defensible as a correct answer in a real pharmacology context — it is not clearly wrong. The question is too abstract and the distractors are not sufficiently differentiated. This is a concept that would be better tested through a specific drug example (e.g., metformin and AMPK, as in benchmark item 67321cbe). Disposition: Fix — anchor to a specific clinical example (e.g., a drug that targets an allosteric site) and rewrite distractors to represent specific mechanistic errors.

660c4d77 — Bence Jones Protein The question ("Which protein chain precipitates at 50°C–60°C but disappears on heating?") tests Bence Jones protein (immunoglobulin light chains), which is a classic clinical biochemistry concept relevant to multiple myeloma. The answer (light chain) is correct. However, the stem does not mention the clinical context (multiple myeloma, proteinuria) and the distractors are minimal. A stronger version would present a patient with bone pain, hypercalcemia, and proteinuria, and ask about the characteristic protein finding in urine. Disposition: Fix — add a multiple myeloma clinical vignette.

9f5eb0f6 — LDH Flipping Effect The concept (LDH1 > LDH2 in myocardial infarction, the "flipping" of the normal LDH2 > LDH1 ratio) is a classic clinical biochemistry PYQ concept (NEET-PG 2018). The question is correctly keyed. However, the stem ("Which of the following indicates the 'flipping effect'?") is a bare recall item. A stronger version would present a patient with chest pain 48 hours after onset and ask which LDH isoenzyme pattern confirms myocardial infarction. The concept is worth keeping; the execution needs a clinical anchor. Disposition: Fix — add a clinical vignette (chest pain, elevated troponin, LDH pattern) and reframe as an application question.

Prioritization

The following prioritization framework is recommended for the content operations team, ordered by urgency and impact.

Tier 1 — Immediate Action Required (Wrong Key / Factually Unsafe) These items must be removed from all active test templates before any further deployment. They are actively harmful.

Question ID	Issue	Action
`16b8f824`	Wrong key: MELAS inhibits Complex I, not Complex II	Disable
`ef7b1ad7`	Wrong key: < 40 mg/dL is HDL threshold, not total cholesterol	Disable
`a4b6d7db`	Wrong key: TPP does not catalyze transamination (that is PLP)	Disable
`ad1b1d0b`	Broken option logic: "both vomiting and renal failure" as NOT-cause answer is self-contradictory	Disable
`209f3c5d`	Wrong key: limiting amino acid in pulses is methionine, not cysteine	Disable or fix key
`324832c9`	Factually unsafe: ammonia-plaque relationship is incorrect; dental content	Disable

Tier 2 — High Priority Fixes (Worthwhile Concept, Weak Execution) These items test high-yield concepts but need rewriting before deployment. They should be queued for the next content sprint.

Question ID	Concept	Required Fix
`ac8aecda`	Porphyria clinical triad	Add enzyme deficiency ask; sharpen distractors
`8f04fb38`	Glucokinase kinetics	Fix distractors; add clinical context
`9f5eb0f6`	LDH flipping effect	Add MI vignette
`660c4d77`	Bence Jones protein	Add myeloma vignette
`13655811`	Pyrimidine metabolism / Reye syndrome	Add Reye syndrome vignette
`52a06b5a`	Mitochondrial metabolic compartmentalization	Sharpen distractor set
`db674f87`	Hawkinsinuria	Add neonatal vignette
`3614e2a9`	Collagen VII anchoring fibrils	Fix broken stem formatting
`2943bcd4`	Allosteric inhibition specificity	Anchor to specific drug example

Tier 3 — Bulk Disable (Low-Value, Correct but Trivial) These items are factually correct but below the quality bar for PG entrance preparation. Given that the benchmark and PYQ sets already cover the same concepts at higher quality, these should be disabled rather than rewritten. The list below is representative, not exhaustive.

e5b8ae37, de76256c, 77b14ab5, ba929cc0, a46ba1ef, 4c48fa72, 172fac46, cdba3715, e23e926f, 8b7f5f78, 6aa10546, cb71eabf, e0e63051, d2153405 (concept covered by better items), 50e5bffd, 0dffd785, 93d8dd7a, 483b278f, 5b5c59c7, 23aca9d9, 2f3edc71, d15441b1, f069ae42, 5475e2d1, 04f06284, ca178c52, 14ac02c4

Tier 4 — Duplicate Resolution a428bffd appears twice (generic and risky sets) — remove one instance and evaluate the remaining item for quality.

Example Keep / Fix / Disable Calls

The following table provides concrete disposition decisions for a representative cross-section of the reviewed sample, with brief rationale.

Question ID	Topic	Disposition	Rationale
`ac8aecda`	Porphyria	FIX	High-yield clinical concept; correct key; needs vignette and sharper distractors
`8f04fb38`	Glucokinase kinetics	FIX	Correct concept; distractors too weak; add clinical context
`9f5eb0f6`	LDH flipping effect	FIX	PYQ concept (NEET-PG 2018); correct key; needs MI vignette
`660c4d77`	Bence Jones protein	FIX	Classic clinical biochemistry; correct key; needs myeloma vignette
`13655811`	Pyrimidine metabolism	FIX	High-yield; correct key; needs Reye syndrome clinical anchor
`3614e2a9`	Collagen VII	FIX	Correct concept and key; stem has formatting artifact (blank placeholder)
`2d2b8cc6`	Urea cycle / hyperammonemia	FIX	Correct and discriminating; add clinical vignette to reach Bloom's 3
`52a06b5a`	Mitochondrial compartmentalization	FIX	Correct concept; distractor set needs sharpening
`db674f87`	Hawkinsinuria	FIX	Unusual but testable IEM; correct key; needs neonatal vignette
`212ab77a`	Signal peptide / secretory proteins	FIX	Correct concept; options are verbose but distinguishable; add ER context
`16b8f824`	MELAS / ETC Complex	DISABLE	Wrong key (Complex I, not Complex II)
`ef7b1ad7`	ATP III cholesterol threshold	DISABLE	Wrong key (< 40 mg/dL is HDL threshold, not total cholesterol)
`a4b6d7db`	TPP and transamination	DISABLE	Wrong key (transamination requires PLP, not TPP)
`ad1b1d0b`	Metabolic alkalosis	DISABLE	Broken option logic; self-contradictory combined option
`324832c9`	Ammonia and dental plaque	DISABLE	Factually unsafe; dental content; not standard Biochemistry curriculum
`4c48fa72`	Lambda phage symbol	DISABLE	Zero biochemical content; tests Greek alphabet recognition only
`e5b8ae37`	Anticodon and tRNA	DISABLE	First-year MBBS level; no PG exam relevance
`a46ba1ef`	Water-soluble vitamins	DISABLE	Pre-clinical recall; no PG exam relevance
`cdba3715`	HMG-CoA reductase	DISABLE	Correct but trivial; concept covered by better gold items
`cb71eabf`	OTC deficiency	DISABLE	Correct but trivial; concept covered by benchmark item `1ebafee0`
`172fac46`	Ig heavy chain chromosome	DISABLE	Wrong subject framing; trivial recall; "all of the above" distractor is logically impossible
`9a2f471d`	Becker MD inheritance	DISABLE	Wrong subject (Pathology/Genetics); no biochemical content
`f71c0096`	Dietary fiber in T2DM	DISABLE	Wrong subject (Nutrition/Medicine); no biochemical mechanism tested
`a428bffd` (duplicate)	Protein disulfide isomerase	DISABLE one instance	Literal duplicate across generic and risky sets; data integrity issue
`883c2bea`	Copper deficiency anemia	DISABLE	Non-specific stem; misleading in isolation; low-yield
`756c4d7e`	HbA1c timeframe	FIX key	Correct concept; key should be "8–12 weeks" or "2–3 months"; current key "6–8 weeks" is imprecise

This report covers the reviewed sample of 100 candidate questions. Findings should not be extrapolated as universal statements about the full 10,646-question pool without further sampling. The patterns observed — particularly the Bloom's compression, the wrong-key cluster in the risky set, and the volume of low-yield trivia — are sufficiently consistent across the sample to warrant systematic remediation rather than item-by-item review.