The literature: every ALS paper, machine-read

Papers per year

Leads: linked to less disease in the literature, not in any trial we hold

Chemicals that at least three papers link to ALS in the “less disease” direction, in any setting from cell dishes to patients, and whose name appears in none of the atlas's trial records. Substances that papers mainly measure in patients, and whole classes of compounds, are left out.

These are leads to check, not findings. The direction was read by software from each abstract and is sometimes wrong. A “less disease” link can mean a reported protective effect, but it can also mean that patients were found to have lower levels of the substance, which is a different thing. Most of this evidence comes from cells and animals, where the great majority of promising results do not carry over to people. Some entries may have been tried under another name, or abandoned for good reasons that an abstract does not show. Open the papers before drawing any conclusion.

Check on the method: the same reading finds what is already in trials

Chemicals read the same way that do match a trial in the atlas. If the method were only noise, approved and trialled treatments would not appear here.

Reported as possible risk factors

Chemicals that papers relate to ALS in the direction of raising risk or worsening it. Many of these are long-debated environmental hypotheses that remain unproven.