Skip to content

SCALAR by SIGMA IV

From spreadsheet to report, with every step visible

SCALAR is desktop software for laboratory researchers. Check your data, ask a question in plain words, and get an answer with its chart, its numbers and the calculation behind it. No programming required.

Try it in your browserRead the guide (PDF)
SCALAR screenshot. The board: each answer is a label, a sentence in plain words, a chart with stars and the numbers.

The board: each answer is a label, a sentence in plain words, a chart with stars and the numbers.

questions in plain language
21questions in plain language
tests, each checked against R
24tests, each checked against R
chart types
31chart types
lab tools
25lab tools

Windows and macOS · English and Dutch · Light and dark theme · Version 4A.11. All screenshots use synthetic data.

Try the workflow

A small, working slice of SCALAR in your browser. Drag a question onto the board, choose the columns and read the answer. Nothing is uploaded.

Interactive demo with synthetic data. It runs four analyses in your browser and is not SCALAR's desktop calculation engine. The numbers match R for this dataset.

  1. Is there a difference between groups?

    IL-6 differs between these pairs after correction: Drug A vs Control, Drug B vs Drug A.

    IL-6 per group, box plot with significanceDrug A vs Control: corrected p 7.03e-7 (****); Drug B vs Drug A: corrected p 2.09e-6 (****); Drug B vs Control: corrected p 3.65e-1 (ns)20406080100120IL-6 (pg/ml)Controln = 12Drug An = 12Drug Bn = 12********ns

    One-way ANOVA: F(2, 33) = 41.90, p < 0.001

    PairDifferencep (Holm)
    Drug A vs Control24.37< 0.001****
    Drug B vs Control2.300.365ns
    Drug B vs Drug A-22.07< 0.001****
    How to read this
    What the chart shows
    The box spans the middle half of the samples (first to third quartile) with the median as a thick line; whiskers reach the smallest and largest value. Brackets mark each pair with its corrected result.
    Data used
    All 36 samples: IL-6 (pg/ml), grouped by treatment (Control, Drug A, Drug B).
    Calculation
    With three groups: a one-way ANOVA, then a Welch t-test for every pair, corrected with Holm's method. With two groups: a single Welch t-test. Stars: **** p < 0.0001, *** < 0.001, ** < 0.01, * < 0.05, ns otherwise.
    How to read it
    A small p-value says the difference would be unusual if the groups were truly the same. It says nothing about why, and nothing about whether the difference matters biologically.
    Limits
    Assumes independent samples. With 12 samples per group, a single outlier can change the result. A clear difference is a reason to repeat the experiment, not a conclusion.
    R code

    Runs on the downloaded CSV and gives the numbers shown above.

    d <- read.csv("scalar-demo-synthetic.csv")
d$treatment <- factor(d$treatment, levels = c("Control", "Drug A", "Drug B"))
summary(aov(IL6_pg_ml ~ treatment, data = d))
pairwise.t.test(d$IL6_pg_ml, d$treatment,
                pool.sd = FALSE, p.adjust.method = "holm")

What SCALAR does

Seven areas in one program. Choose one to see the real interface.

SCALAR screenshot. Step 1: the files of a project and a first look at the rows.

Step 1: the files of a project and a first look at the rows.

Bring a spreadsheet, keep the original

Drop a CSV or Excel file. SCALAR checks it first and never changes your file. Then drag each column into a role: measurement, group, donor or time point.

  • A first look at every file of a project
  • New columns such as log2, fold of control or qPCR ΔΔCt
  • Questions are only offered when the columns can answer them

Turn a surface yourself

A browser version of SCALAR's 3D surfaces. Drag to turn, zoom, change the colour map and read any point.

Interactive demo with synthetic data, drawn in your browser. SCALAR's own 3D views are shown in the screenshots above.

Every number can be traced

Results you can check are more useful than results you have to trust.

  1. The full calculation

    Each answer can show every step: the formula, the numbers filled in and the result.

  2. An Excel workbook

    A workbook that recomputes the answer, so a colleague can follow it without SCALAR.

  3. R code with the data

    A script that runs on its own and gives the same result. Each test is checked against R.

  4. A report in one click

    A whole board becomes a PDF: summary, method per question, charts and an appendix with every number.

Your data stays on your computer

SCALAR has no cloud component. Files, models and reports are processed and stored locally and nothing is uploaded, so unpublished data stays under your control.

Good to know

  • SCALAR is exploratory. A clear pattern is a reason to repeat an experiment, not a conclusion.
  • It is research software and does not make medical diagnoses.
  • It covers a broad set of common laboratory analyses and hands you R code to continue. It does not replace every feature of R or GraphPad.
  • Not included: alignment of raw sequencing reads, neural networks on images, protein design, and downloading gene sets.

Availability

SCALAR is not yet available for public download. Release files and licensing will be published here when they are ready. The guide already describes everything version 4A.11 can do, with a synthetic dataset and a worked example for every test and tool.

Download the guidePDF, 28 pages, 3.8 MB