Biochemistry Editing and Proofreading Services

A figure shows three bars with error bars and an asterisk. Somebody trying to judge whether it means anything needs to know how many independent experiments it represents, whether the replicates were biological or technical, what the error bars are, and what test produced the asterisk — and none of that is in the figure, the legend or the methods, because it was all obvious to the person who did the work.

We edit what biochemists write — journal articles and their figures and legends, methods sections and their reproducibility, assay and enzyme kinetics reporting, protein purification and characterisation sections, buffer, reagent and antibody documentation, structural and biophysical results sections, supplementary information and data availability statements, grant applications and progress reports, and thesis chapters and defence material. Our editors work on the figure legend that has to carry the experiment.

The figure legend is where biochemical evidence is either assessable or taken on trust, and its failure is a caption that names the axes. A reader cannot evaluate a result whose replication structure is invisible. We work through these so every legend states the number of independent biological replicates, and separately the technical replicates within them, since three wells of one preparation and three separate preparations are entirely different evidence and the field's reproducibility problem lives in that distinction; so the error bars are identified as what they are — standard deviation, standard error or confidence interval — because they differ by a factor that changes the visual impression completely; so the statistical test is named with what was compared and whether correction was applied, given that an asterisk with no test behind it is decoration; so the exact sample size appears as a number rather than a range, as "n = 3–5" means the reader cannot reconstruct any comparison; so the reagent identity is given with enough detail to repeat it — the antibody's supplier and catalogue number, the protein's expression system and tag, the buffer's exact composition — since these are the variables that most often explain a failed replication; so anything normalised is stated with what it was normalised to; and so representative images are declared as representative with how many were examined. Legends written this way let a reader judge the experiment.

Everything you send is treated in confidence, including unpublished data, figures and manuscripts. We are editors rather than biochemists or statisticians, and we offer no view on your experiments, assays or results. What we can do is make the legend carry what the reader needs to judge the figure.