Create publication-ready forest plots from effect-size and confidence-interval data.
Upload study labels, effect estimates, and interval bounds, describe subgroups or summary rows, then generate a clean AI-made forest plot with manuscript-ready PNG output.
Upload data and describe the AI chart you need.
Costs 5 credits. Uses your uploaded data and prompt to generate a publication-ready PNG.
Preview
Aspect ratio: 16:9
These are sample previews. Upload your dataset to generate your own forest plot.
Use these examples as visual directions for meta-analysis, subgroup effects, adjusted models, sensitivity checks, weighted studies, and manuscript-ready forest plots.
Prompt
Based on the uploaded treatment_effect_meta_analysis.csv file, create a publication-ready forest plot comparing treatment effects across studies. Use the study label column, plot effect estimates with 95% confidence intervals, add a no-effect reference line at 1.0 for ratio measures, and include a pooled summary diamond only if the uploaded data contains an overall or pooled row.
Prompt
Based on the uploaded diagnostic_odds_ratio_by_cohort.xlsx file, create a clean forest plot for diagnostic odds ratios. Use a logarithmic odds-ratio x-axis when appropriate, show horizontal 95% confidence intervals, size markers by weight if a weight column is supplied, and keep the right-side estimate and weight columns compact.
Prompt
Based on the uploaded subgroup_treatment_effects.csv file, create a subgroup forest plot. Group rows by the uploaded subgroup column, show study estimates and confidence intervals within each subgroup, add subgroup summary rows only when provided, and use restrained blue and green accents with clear study labels.
Prompt
Based on the uploaded adjusted_model_coefficients.tsv file, create a forest plot for adjusted model coefficients. Use predictor names as row labels, plot standardized beta estimates with 95% confidence intervals, add a no-effect reference line at 0.0, color negative and positive effects subtly, and do not add a pooled effect diamond.
Prompt
Based on the uploaded sensitivity_analysis_intervals.csv file, create a sensitivity-analysis forest plot. Show each analysis scenario as a row, plot mean differences with lower and upper confidence bounds, add a no-effect reference line at 0.0, include a subtle reference band only if requested or supplied, and keep the figure clean for a manuscript.
Prompt
Based on the uploaded weighted_study_comparison.xlsx file, create a weighted study forest plot. Use study labels, odds ratio estimates, lower and upper confidence intervals, and weight values from the uploaded columns; vary marker size by weight, add a no-effect line at 1.0, and include an overall row only when it exists in the file.
Add CSV, TSV, TXT, or Excel data with row labels, effect estimates, lower bounds, and upper bounds.
Tell SciFigure AI whether the measure is a ratio, difference, coefficient, subgroup analysis, or model-result figure.
Create a clean AI forest plot, verify intervals and reference lines, then download PNG for papers and slides.
It turns study-level estimates, model coefficients, odds ratios, risk ratios, hazard ratios, mean differences, or subgroup effects into a clean forest plot with row labels and confidence intervals. The result is designed for manuscripts, lab reports, systematic reviews, posters, and research presentations.
Use one label column, one numeric effect estimate column, and numeric lower and upper confidence interval columns. Optional subgroup, weight, and sample-size columns can guide layout.
Yes. Upload a subgroup column or describe the subgroup rows in your instructions, then verify the generated grouping before using the figure.
Yes. Ratio measures usually use a reference line at 1.0, while mean differences, beta coefficients, and other signed estimates usually use 0.0.
Only ask for a pooled or overall diamond when that row exists in the uploaded data or the prompt provides the summary estimate and interval.
Yes, but verify the uploaded estimates, interval bounds, scale, labels, weights, and statistical interpretation before submission.
Generate forest plots, error bar charts, box plots, and other research-ready visualizations with SciFigure AI.
