LT50 Calculator

v1.1 · 2026.11.10

This tool estimates LT50, the temperature that kills 50% of plants or tissues, as a measure of freezing tolerance. Copy temperatures and survival (or electrolyte leakage) from a spreadsheet and paste them in: a sigmoid curve is fitted, LT50 is calculated and a graph is drawn. Add headings such as genotype or treatment to fit and overlay each group separately.

Your data is never uploaded: all calculations run in your browser.

Data (paste straight from a spreadsheet)

Accepted layouts

Wide and long layouts are detected automatically.

Wide: temperature in the first column and values in the following columns. Any number of heading rows (genotype, treatment, …) is allowed, and columns with the same heading are treated as replicates. Merged cells and replicate rows such as “1, 2, 3” or “rep1” can be pasted as they are. Blank cells are treated as missing, so groups may differ in N and in temperature points.

Long: one observation per row — temperature, value, then group names (any number of columns from the third).

Curve asymptotesFixed at 0% and 100% (for survival)Estimated from the data (for electrolyte leakage and other data not spanning 0–100%)

Definition of LT50Midpoint of the curve (inflection point)Temperature at which the value is exactly 50%

X-axis labelY-axis label

Graph displayColder temperatures to the rightShow LT50 values on the graph

Paste data to see the fitted curves here.

Method

You may use the following description in the methods section of a paper.

A four-parameter logistic function, y = L + (U − L) / (1 + exp(−k (T − T_m))), was fitted by nonlinear least squares using the Levenberg–Marquardt algorithm, where L and U are the lower and upper asymptotes, k is the slope and T_m is the temperature at the inflection point. When the asymptotes were fixed (L = 0, U = 100), T_m was taken as LT50. When they were estimated from the data, LT50 was taken as either T_m or the temperature at which the fitted curve crosses 50%. The standard error of LT50 was obtained by the delta method, and the 95% confidence interval was calculated from the t distribution with n − p degrees of freedom (n, number of data points; p, number of estimated parameters). Calculations were performed with LT50 Calculator version 1.1 (Takahashi, 2026).

Results were checked against fits in R (the nls function): LT50 and its standard error agree to four decimal places.

How to cite

If you use results from this tool in a paper or presentation, please cite it as follows. Replace the access date with the date you used it.

Takahashi D (2026) LT50 Calculator, version 1.1. Laboratory of Stress and Adaptation Biology, Saitama University. https://www.adaptationbiology.com/en/tools/lt50/ (accessed 2026-11-10)

Results copied with “Copy results” also record the version used.

Disclaimer

This tool is provided free of charge, as is, to support research. We make no warranty as to the accuracy of the results or their fitness for any particular purpose; please verify results yourself before relying on them. The author and the Laboratory of Stress and Adaptation Biology, Saitama University, accept no liability for any loss or damage arising from the use of this tool. The tool may change without notice.

v1.1: 2026.11.10　Improved graph display (per-series colours, toggle for LT50 values, framed axes with ticks, choice of temperature direction). Calculations unchanged
v1.0: 2026.11.10　First release

All tools →