The evidence circuit that closes the outbreak investigation. Five tests, five classes of agent, one name at the end.
By the end of this block you will name the class of agent causing this outbreak, then name the organism itself, and you will be able to say which piece of evidence earned each of those two claims.
Time: 80 minutes, in class. The wet lab is timed. Everything else is at your own pace.
Infectious disease still kills millions of people a year worldwide, and the single largest reason a treatable infection becomes a fatal one is time lost naming it. A doctor who does not know the class of agent is choosing a drug by guess, and a guess against the wrong class does nothing at all.
An antibiotic does nothing to a virus. An antifungal does nothing to a bacterium. Before anyone can choose a treatment, somebody has to answer a narrowing question: what kind of thing is this?
There are five classes on the table today. Each mini lab removes some of them. By SEC 06 you should be down to one class, and BLAST gives you the organism inside it.
The reference table of agent characteristics is in the index at the back of this packet. It is given to you, it is not something you derive today, and its sources are printed with it. Turn to it whenever a mini lab asks you to rule a class in or out.
Before any evidence: write which class you would guess, and why. You will check this against your own result in SEC 07.
Epidemiology is the part of medicine that works on populations rather than one patient, and the shape of an outbreak curve has redirected real investigations. One sharp peak says everybody met the same thing once. A rolling series of peaks says people are giving it to each other, and those two findings call for completely different responses.
You have two pieces of population evidence: the patient record in your evidence portfolio, and the outbreak curve below. Neither names an organism. Both narrow the field.
| Agent class | Still possible after the curve?yes / no | The evidence that decided it |
|---|---|---|
| Bacteria | ||
| Viruses | ||
| Fungi | ||
| Protozoa | ||
| Parasitic worms |
Does this curve show one shared exposure or person to person spread? Give the feature of the shape that told you.
Which classes did you rule out on onset speed alone, and which stayed in?
Every blood test that reports a number rather than a yes or no depends on a standard curve, and a standard curve is built from a dilution series. The series you make in the next twenty minutes is the one tomorrow's ELISA is read against, so a careless tube today becomes a wrong concentration tomorrow.
This is the only wet lab today, and all three of my classes run it, so the bench is set the same way in every period.
A serial dilution is the same dilution repeated. Each tube takes 1.0 mL from the tube before it into 9.0 mL of water, so each tube is one tenth as concentrated as the one before. Five steps takes you from the stock to one hundred thousandth of it.
You are using hydrochloric acid and the indicator phenolphthalein. The indicator is colourless in acid and pink above pH 8.2, so counting the drops of base needed to turn each tube pink measures how much acid is in it. That drop count against dilution step is your standard curve.
Two control tubes are already made up in your rack.
Check both before tube 1 and again at the end, and record both checks. If C1 is not pink, stop.
| Tube | Dilutionfactor | Concentrationmol/L | Drops to reach pinkdrops |
|---|---|---|---|
| C1 | ('control',) | ('0.1 M base',) | |
| C2 | ('control',) | ('water only',) | |
| 1 | ('1:10',) | ('0.010',) | |
| 2 | ('1:100',) | ('0.0010',) | |
| 3 | ('1:1,000',) | ('0.00010',) | |
| 4 | ('1:10,000',) | ('0.000010',) | |
| 5 | ('1:100,000',) | ('0.0000010',) |
Each step is ten times weaker. Describe what happened to the drop count at each step.
Which was the last tube where you could still detect the acid? That is your detection limit.
Why can a standard curve tell you the concentration of an unknown sample? Answer in one sentence.
The Gram stain is over 140 years old, costs almost nothing, and is still run within minutes of a serious infection arriving in a hospital, because it splits the bacterial world in two and that split changes the first antibiotic given. Few tests in medicine buy that much for that little.
A Gram stain answers two questions at once: is anything there, and what shape and stain type is it. Purple means the cell held the first stain (Gram positive). Pink means it did not and took the counterstain instead (Gram negative). Round cells are cocci. Rod-shaped cells are bacilli.
The culture answers a different question: will it grow on ordinary agar. That single result separates bacteria and fungi from viruses, protozoa and worms.
Two controls, each ruling out a different failure.
| Observation | Your result | What it rules out |
|---|---|---|
| Stain control passed? | ('if no, stop',) | |
| Sterile plate stayed clear? | ('if no, stop',) | |
| Gram reaction of the isolate | ||
| Cell shape | ||
| Grew on standard agar? |
After the stain and the culture, which single class of agent is still standing? Name the two results that removed the others.
A classmate says the stain proves which species it is. Write the one sentence you would say back.
Running a patient sample against a known ladder is how a laboratory turns an invisible molecule into a measurement anyone else can check. The same method sizes a virus fragment, confirms a bacterial gene, and reads a DNA profile in a criminal case. One idea, used three ways.
The laboratory ran PCR on the isolate using primers for a gene region that differs between bacterial species, then ran the products on a gel. Short fragments travel further than long ones, so the distance a band travelled tells you its size once you have a ladder to read against.
This gel narrows the candidates. It does not name one. That is the next mini lab.
The same two control lanes as any run: a no-template lane that must stay empty, and a positive control that must show its band. Check both before reading anything else.
| Lane | Distance travelledmm | Size from your curvebp | Same size as the patient?yes / no |
|---|---|---|---|
| Patient isolate | |||
| Candidate A | |||
| Candidate B | |||
| Candidate C | |||
| Candidate D |
How many candidates still match the patient band, and which are they?
You have more than one candidate left with the same band size. Write the one sentence that says what this gel can and cannot establish.
Sequencing turned identification from a week of culturing into an afternoon, and public databases mean a school in Cleveland runs the same search a reference laboratory runs. The read below is a real 16S sequence and the tool is the real tool. Nobody in this room, including me, is going to tell you the answer, because the whole point is that the evidence tells you.
The 16S rRNA gene is carried by every bacterium. It is similar within a species and different between species, which is what makes it the standard gene for naming a bacterium from its DNA.
Below is one 500 base read from the patient isolate. The name of this organism appears nowhere in this packet, nowhere in your portfolio, and nowhere on the portal. You are going to get it from BLAST.
The three number rule, which you already used on Monday. Read all three together before you name anything.
Your isolate read. Copy the whole thing, including the line that starts with the greater-than sign.
>S2_practice_isolate 16S rRNA gene, one read of 500 bases GGCTCAGGATGAACGCTGGCGGCGTGCCTAATACATGCAAGTCGAGCGAACGGACGAGAAGCTTGCTTCTCTGATGTTAGCGGCGGACGGGTGAGTAACACGTGGATAACCTACCTATAAGACTGGGATAACTTCGGGAAACCGGAGCTAATACCGGATAATATTTTGAACCGCATGGTTCAAAAGTGAAAGACGGTCTTGCTGTCACTTATAGATGGATCCGCGCTGCATTAGCTAGTTGGTAAGGTAACGGCTTACCAAGGCAACGATGCATAGCCGACCTGAGAGGGTGATCGGCCACACTGGAACTGAGACACGGTCCAGACTCCTACGGGAGGCAGCAGTAGGGAATCTTCCGCAATGGGCGAAAGCCTGACGGAGCAACGCCGCGTGAGTGATGAAGGTCTTCGGATCGTAAAACTCTGTTATTAGGGAAGAACATATGTGTAAGTAACTGTGCACATCTTGACGGTACCTAATCAGAAAGCCACGGCTAACTA
| Scientific name | Accession | Identity% | Query cover% | E value | |
|---|---|---|---|---|---|
| Top hit | |||||
| Runner-up, different species |
Your prediction, written before the results appeared:
Your identification sentence. Name the organism and cite all three numbers inside the sentence.
The claim ceiling: a BLAST match names the organism your read most resembles in the database. Write what that does not establish about this patient.
Naming an organism is only useful if it changes what happens to the patient, and it does. The same infection treated with the right drug and the wrong drug produces two different outcomes, and for at least one organism on the table below the reflex answer, an antibiotic, is the wrong answer.
Turn to the treatment and prognosis table in the index. Find your organism, or the closest candidates if your three numbers did not settle it.
Read the whole row. The treatment column and the prognosis column are different questions, and for two organisms in that table the prognosis depends on a laboratory result that has not been run yet.
The organism your evidence supports, and how confident you are in one word:
The treatment its row gives, copied exactly:
One thing about the prognosis that depends on a test nobody has run yet. Read the last row of the patient record before you answer:
Go back to SEC 01 and read your first guess. Was it right? Write which piece of evidence would have changed your mind earliest.
Extra credit, at home, on the separate sheet. Write the treatment plan you would give as the physician on this case: the diagnosis, the drug and why, what you would monitor, what would tell you that you were wrong, and what you would tell the patient in plain words. It is optional and it is graded on reasoning, not on getting the same answer as anybody else.
| Class | What it is | Size | Seen on a Gram stain? | Grows on standard agar? | What it responds to | Typical onset | Example illnesses |
|---|---|---|---|---|---|---|---|
| Bacteria | One cell, no nucleus (prokaryote) | 0.5 to 5 micrometres | Yes, Gram stain shows them | Yes, on standard agar | Antibiotics, chosen by susceptibility | Hours to a few days | Strep throat, listeriosis, tuberculosis |
| Viruses | Genetic material in a protein coat, not a cell | 20 to 300 nanometres | No, far too small for a light microscope | No, they need living host cells | Antivirals. Antibiotics do nothing | One to fourteen days | Influenza, measles, COVID-19 |
| Fungi | Cells with a nucleus and a chitin cell wall | 2 to 10 micrometres (yeasts), longer threads (moulds) | Poorly. A KOH prep or a fungal stain is used | Yes, on fungal media such as Sabouraud agar | Antifungals | Days to weeks | Thrush, athlete's foot, histoplasmosis |
| Protozoa | One cell, with a nucleus (eukaryote) | 10 to 50 micrometres | Not by Gram. Seen on a stained smear or wet mount | Special media only, often not cultured at all | Antiprotozoal drugs | Days to weeks | Malaria, giardiasis, amoebic dysentery |
| Parasitic worms | Many cells, a whole animal (helminth) | Millimetres to metres | No. Eggs are found in stool under a microscope | Not cultured | Antiparasitic (antihelminthic) drugs | Weeks to months | Tapeworm, hookworm, roundworm |
| Organism | Class | First-line treatment | Prognosis and what it depends on | How it usually reaches people |
|---|---|---|---|---|
| Listeria monocytogenes | Bacterium | Ampicillin, often with gentamicin for severe or invasive disease | Dangerous in pregnancy, in newborns, over 65, and with a weakened immune system. Invasive disease is serious and treated in hospital. | Food-borne, including deli meats, soft cheeses and unwashed produce |
| Staphylococcus aureus | Bacterium | Depends entirely on the susceptibility test. Methicillin-sensitive strains take a beta-lactam such as cefazolin; MRSA does not, and is treated with drugs such as vancomycin | Ranges from a minor skin infection to life-threatening bloodstream infection. The susceptibility result changes the drug, so the lab result changes the prognosis. | Skin and nose carriage, wounds, healthcare contact |
| Streptococcus pyogenes | Bacterium | Penicillin or amoxicillin. Still reliably sensitive | Good with treatment. Untreated infection can lead to rheumatic fever, which is why a sore throat gets cultured. | Person to person by droplets |
| Pseudomonas aeruginosa | Bacterium | An antipseudomonal drug such as piperacillin-tazobactam or ceftazidime. Many ordinary antibiotics do not work | Often affects people already in hospital or with weakened defences. Resistance is common, so treatment follows the susceptibility test. | Water, moist surfaces and equipment |
| Salmonella enterica | Bacterium | Usually fluids and supportive care. Antibiotics are reserved for severe or invasive illness | Most healthy people recover. Antibiotics are not automatic, because they are not always helpful. | Food-borne, including poultry, eggs and produce |
| Escherichia coli, Shiga toxin-producing | Bacterium | Fluids and supportive care. Antibiotics are generally avoided, because they may raise the risk of a serious kidney complication | A case where the reflex answer is the wrong answer. This is why naming the organism matters before choosing a drug. | Food-borne, including undercooked beef and produce |
| Bacillus subtilis | Bacterium | Rarely causes disease in healthy people. Usually treated as an environmental or laboratory contaminant | A hit on this organism is more often a contamination question than a diagnosis. Check your controls before you check the drug chart. | Soil, dust and laboratory surfaces |
Claim ceiling. This table is a classroom reference for naming and reasoning. It is not a prescribing guide, it is not medical advice, and no one in this room diagnoses or treats anybody.
Table 1 is standard microbiology: cell structure, size ranges, staining behaviour, culture requirements and drug classes as described by the CDC and the National Library of Medicine. Table 2 follows CDC treatment guidance and standard clinical references for each named organism. Both are printed here as reference, not derived by students, which is why they sit in the index and not in a work section.
The isolate read in SEC 06 is a real 16S rRNA sequence. The identification it produces is produced by you, in BLAST, on the day. It is recorded nowhere in this packet.