Second cohort · 2026
Thursdays and Saturdays, 5–7 pm.
Seventeen live sessions on Google Meet, 3 September to 31 October. Each one comes with a single video or reading — do it before you arrive and the session will make far more sense.
The year ahead
Everything Basira is running to the end of 2027Two cohorts a year, both in windows where students have no exams. Between them, a two-week crash course ending in a competition — write the best hypothesis and preliminary abstract, and the prize is carrying that work into a real paper with a mentor.
Remind me
One email when applications open for the thing you picked. Nothing else, and you can say stop at any point.
Second cohort, session by session
Chapter 1 · 1·0–1.5
What research is · the claim ladder · strong inference · the two errors
5:00 Aroob — the chapter · 6:05 Breakouts, 15 each — Stage 0 — accounts, repository and tools working
Chapter 1 · 1.6–1.11
Reading a result · the four traps · calibrated language · abduction
5:00 Haneen — the chapter · 6:05 Breakouts, 15 each — Stage 0 — finish setup; Assignment 1 opened together
Chapter 2 · 2·0–2.6
Integrity · the bright lines · the grey zone · structure over willpower
5:00 Aroob — the chapter · 6:05 Breakouts, 15 each — Stage 1 — your question, first draft
Chapter 2 · 2.7–2.12
How a group works · critique · shared code · misleading figures
5:00 Haneen — the chapter · 6:05 Breakouts, 15 each — Stage 1 — pairing for the Assignment 2 critique
Chapter 3 · 3·0–3.4
The cell · DNA to protein · mutations · accelerators and brakes
5:00 Aroob — the chapter · 6:05 Breakouts, 15 each — Stage 2 — start the field map
Chapter 3 · 3.5–3.7
Cancer as evolution · cancer as many diseases · the hallmarks
5:00 Haneen — the chapter · 6:05 Breakouts, 15 each — Stage 2 — the map, continued
No session — National Day holiday
Chapter 3 · 3.8–3.9
Reading a tissue section · a tumour is a tissue, not a bag of cancer cells
5:00 Aroob — the chapter · 6:05 Breakouts, 15 each — Stage 2 — last clinic before it is due Tue 29 Sep
Chapter 3 · 3.10–3.14
Metabolism · the pathways · the mitochondrion · the Warburg effect
5:00 Haneen — the chapter · 6:05 Breakouts, 15 each — Stage 3 — nearest papers side by side
Chapter 3 · 3.15
Where the open questions live · four ideas this group abandoned
5:00 Aroob — the chapter · 6:05 Breakouts, 15 each — Stage 3 — the untraced claim log, Assignment 3
Chapter 4 · 4·0–4.5
Four data types · the resources · identifiers · the two silent killers
5:00 Haneen — the chapter · 6:05 Breakouts, 15 each — Stage 4 — aim: one sentence, one null
Chapter 4 · 4.6–4.10
The toolkit · reproducibility · defensive loading · batch effects
5:00 Aroob — the chapter · 6:05 Breakouts, 15 each — Stage 4 — the data audit, Assignment 4
Chapter 5 · 5·0–5.4
Using the reference · describing a column · the z-score · distributions
5:00 Haneen — the chapter · 6:05 Breakouts, 15 each — Stage 5 — pre-specification, first draft
Chapter 5 · 5.5–5.7
The logic of a test · what a p-value is not · effect size · multiple testing
5:00 Aroob — the chapter · 6:05 Breakouts, 15 each — Stage 5 — the refuting sentence
Chapter 5 · 5.8–5.9
Regression and adjustment · screen versus arbiter · honesty in code
5:00 Haneen — the chapter · 6:05 Breakouts, 15 each — Stage 5 — Assignment 5, the design
Chapter 5 · 5.10–5.15
Working the reference — choosing a method for a question you have
5:00 Aroob — the chapter · 6:05 Breakouts, 15 each — Stage 5 — final clinic before it is due Tue 27 Oct
Chapter 6 · 6.1–6.4
Find a question · sharpen it · check it is worth doing · pre-register
5:00 Haneen — the chapter · 6:05 Breakouts, 15 each — Stage 6 — run it, first pass
Chapter 6 · 6.5–6.8
Run it · read the result · decide what happens next · presentations
5:00 Aroob — the chapter · 6:05 Breakouts, 15 each — Presentations, and handover to the project period
After the course — the project period
Teaching stops on 31 October. The project does not. Five supervision meetings carry each project from a sharpened hypothesis to a written paper, in groups of three. Two of them happen while term is still running, because a hypothesis tightened in November is a hypothesis tightened too late.
The cohort splits in half. You have one supervisor from October to December — the same person who runs your breakout in the workshop half of every session, and the same person you see in all five meetings.
Tighten the hypothesis
One sentence, no clauses joined by “and”. The null written down. A positive control that has to fire.
Out of it: Stage 4 signed off
Write the method
What will be run, what counts as a positive result, what counts as a null — all of it before any data is opened.
Out of it: Stage 5 filed. It cannot be revised afterwards.
Run it
The positive control fires before anything else is believed. Every deviation from the pre-registration logged with its reason.
Out of it: Stage 6, analysis complete
Read the result
What it supports, what it does not, and what could not be tested. A null is a result and is written as one.
Out of it: the interpretation, agreed
Write it, and decide where it lands
Full draft, and a named target — a journal, a preprint server, or a conference with its deadline.
Out of it: Stage 7 and the destination
30 November
Every project fully drafted. Not polished — drafted, with its figures, its methods and its limits written.
17 December
Submitted. Sending it is an afternoon, not a fortnight, which is why it can sit inside the exam period.
Groups of three
Ten groups across the cohort, five to each supervisor. One to one would be a hundred and fifty meetings, which is not a thing that can happen.
Exams, 6–31 December
Basira is closed. Nothing is asked of anyone in that window except the act of pressing submit.
Authorship, decided now rather than in December
If a project becomes a paper, the student who did the work is first author. Whoever supervised it appears as a co-author. Nobody who did not contribute goes on it, and that includes us.
If a student stops replying for three weeks, the project pauses and their name stays on whatever they built. It does not transfer to someone else and it does not get finished without them.
After 31 October
And then it keeps going.
The taught sessions end on 31 October. The capstone does not — Assignment 6 continues as supervised project work through November, finishing with a presentation once your analysis has actually been run.
Designing, running and interpreting research takes longer than a two-hour session, and the schedule reflects that.
Download the full plan