The sentence that appeared most often in the review comments was: What do the error bars in Figure 3 represent? I revised the figures four times, and only on the last pass did I understand: half of paper figure and table standards is about “data to figure,” which tools can help with; the other half is about “figure to conclusion,” which tools cannot help with.
1. Data first becomes a table, then a figure.
All my figures start from one data table: what quantity is on the x-axis, what quantity is on the y-axis, whether each point is a mean, how the error is calculated, and how many samples there are. If the table is incomplete, the figure will inevitably lack information. I used MuchSci’s Office Assistant Agent to organize this table into an editable Excel file (describe the need in one sentence, and it produces editable Word / Excel / PPT), filling in column names, units, and sample counts clearly in one go; later, each time I revise a figure, I only change this one table and do not redraw it. Units in figures and tables must be written correctly in the table header; do not add them temporarily on the figure.
2. Figure captions must be understandable on their own.
The standard is: if the caption is extracted on its own, readers can know what is plotted without reading the main text. It must clearly state what the symbols mean, what the error bars represent (standard deviation, standard error, or confidence interval; journal requirements differ), the number of replicates, and the statistical test. If the target is an English-language journal, I first use MuchSci’s Paper Translation to produce a version of the caption (bidirectional Chinese-English academic translation, preserving formula and terminology consistency), then check it sentence by sentence—terminology in the caption being inconsistent with the main text is the easiest thing for reviewers to spot at a glance.
3. The “not involved” parts are here.
· What the error bars represent: standard error and standard deviation differ by severalfold, and how the caption is written depends on your statistical definition. This is determined by experimental design and statistics; the tool does not know which one you calculated.
· Whether group A and group B can be plotted together: depends on whether the two groups are from the same batch and under the same conditions. This can only be judged by yourself.
· Whether the trend can be stated this way: connecting four points into a line and saying “monotonically increases with rising temperature” is overinterpretation. The figure can be drawn, but the conclusion cannot be stated that way.
· Which figure should go in the main text and which in the supplementary materials: depends on whether it supports the main argument or corroborating evidence.
· Pixel-level specifications: line width, font size, dpi, vector or bitmap, colorblind-friendly color schemes—these are hard requirements of the journal and must be adjusted according to the Author Guidelines; tools cannot replace that.
4. Figure version management.
A figure goes through “draft sketch → first draft → revision version → final version,” and file names do not tell the truth on their own. I store the vector source file and exported image for each figure together in MuchSci’s File Asset Library (centralized storage for generated documents, images, and text, with preview and create/read/update/delete support), and write the data definition marked on the figure and the corresponding data table into a text note; when showing figures to my advisor, I send them via “My Share” so they do not take up chat history.
5. Where the data in the figures comes from.
For calculations involving compound descriptors and molecular property prediction, I use MuchSci’s Cheminformatics Agent (toolchain includes RDKit, Datamol, PyTDC, MedChem, and TorchDrug, and can perform molecular operations, descriptor calculations, and GNN prediction). During its run, I can see progress and interrupt at any time; at key decision points it pops up decision cards for me to confirm—this matters more to me than “producing a pile of numbers”: I need to know how each number was calculated before I dare to mark it on a figure. The calculated results and the original literature used for verification are archived together in MuchSci’s Personal Knowledge Base (supports PDF, DOCX, PPTX, XLSX, TXT, and MD; personal use by default, not shared).
Three boundaries: figures must not beautify conclusions; erasing outlier points and truncating error bars are not formatting issues but academic integrity issues; color schemes and axis ranges affect readers’ judgment; shrinking the vertical axis can make changes look more obvious, and I will not do that; results from computational tools must be reproducible; record parameters and versions together, otherwise six months later even you will not be able to explain clearly how that figure was made.
In the figure and table step, this kind of graduate student AI assistant does the following: organizes data into tables, translates figure captions smoothly, and manages files well. How that line on the figure is drawn is still my own decision.