SciDAVis for Life Sciences & Pharma: Dose–Response, Kinetics, and Descriptive Stats

Plate readers, enzyme assays, cell counts, growth curves — life-science data is mostly small tables and big questions: what is the IC50, does this enzyme follow Michaelis–Menten kinetics, are these groups really different. SciDAVis (free, open-source; latest release v2.9.2, precompiled binaries on SourceForge are v2.8) covers the core of wet-lab statistics: sigmoidal dose–response fits, custom kinetic models, histograms and box plots, and one-click descriptive statistics. This guide shows each workflow and states plainly where the tool stops.

Version note: written for v2.9.x (SourceForge binaries are 2.8). The statistics and fitting commands described here are unchanged between the two.

Step 1 — Dose–response curves and IC50/EC50

Enter concentration and response, plot the curve, then Analysis → Quick Fit → Fit Boltzmann (Sigmoidal). The fitted sigmoid’s midpoint parameter is the EC50 (or IC50 for inhibition) concentration — read it from the fit log together with its standard error. Practical notes: plot against log(concentration) when the doses span orders of magnitude; the fit handles the scale either way. Use Graph → Add Error Bars… to show the spread of replicates — a dose–response without replicate error bars is a claim without evidence. The slope parameter of the sigmoid (Hill slope) is worth reporting when comparing compounds — a slope far from 1 hints at assay artifacts or cooperative binding.

Then check the residuals by eye: zoom the graph and confirm they scatter randomly around zero. A systematic wave means the response is not sigmoidal (biphasic response, partial agonism) and no single IC50 is meaningful. The fitting mechanics, including how to read the results log and judge a fit, are in the curve fitting tutorial.

Screening several compounds? Fit each dose–response in its own table, then collect the midpoints and standard errors into a summary table — one row per compound — for a clean potency ranking. Keep the raw curves in the project: a ranked table without the underlying fits is not auditable, and reviewers will ask.

Step 2 — Enzyme kinetics without a built-in model

Honest statement first: SciDAVis has no built-in Michaelis–Menten fit. The route is Analysis → Fit Wizard (Ctrl+Y) with a custom formula — for example v = Vmax*x/(Km+x) — where you name the parameters, supply starting guesses, and optionally restrict the x-range. Good starting guesses come free from the data: Vmax near the highest observed rate, Km near the substrate concentration giving half that rate. The fit returns Vmax and Km with standard errors, and it is the statistically honest route: fit the raw data, not the linearized Lineweaver–Burk transform (linearization distorts the error structure, so its confidence intervals mislead). If a reviewer asks for the double-reciprocal plot anyway, make it as a display figure computed from the fitted parameters — not as the fitting route.

Residual checks apply here too: curvature in the residuals of a Michaelis–Menten fit is how you discover substrate inhibition or cooperativity — the model failing is data, not an error.

Step 3 — See the distribution: histograms and box plots

Before any test, look at the data. Select your data columns in the table, then Plot → Statistical graphs → Histogram builds a frequency histogram — the default binning uses 10 steps between the min and max of the Y-values and can be adjusted in the Custom curves dialog — ideal for cell-size distributions or replicate spread. Plot → Statistical graphs → Box Plot summarizes each column as a box (mean, min/max, 5th and 95th percentiles), which is the fastest honest comparison of treated versus control groups. If the box plots of two groups barely overlap, you have an effect worth testing properly; if they overlap completely, no test will save the story. Comparing several columns at once works too: Statistical graphs → Stacked Histogram draws one histogram per layer in vertically stacked panels, handy for scanning replicate batches for an outlier run.

Step 4 — Descriptive statistics in one click — and what is missing

Select the column(s) in the table and choose Analysis → Statistics on Columns. SciDAVis creates a new table, “Column Statistics of <your table>”, with one row per column: mean, standard deviation, variance, min, max, N, and more. Analysis → Statistics on Rows does the same across rows, useful when each row is one experimental replicate. Only numeric columns are summarized; empty or invalid cells are ignored rather than treated as zeros — the N column tells you how many valid values were actually used, so check it before quoting any mean. This is the entire built-in statistical capability of the program — and here is the honest boundary: SciDAVis 2.9.x has no t-test, no ANOVA, and no normality test. For those, export the table as CSV and run them in Python/SciPy (SciDAVis has a documented Python scripting API for the repetitive parts), R, or JASP — all free. Do not describe p-values from a tool that cannot compute them. The statistics tutorial documents exactly what the statistics commands produce and the workarounds for the missing tests.

SciDAVis for Chemistry & Spectroscopy: Honest Peak Fitting

UV-Vis, FTIR, and NMR spectra arrive as columns of wavelength (or wavenumber, or ppm) versus intensity. The honest middle of spectroscopy work — importing the instrument export, dealing with the baseline, deconvolving overlapping bands, building a calibration curve — is exactly what SciDAVis (free, open-source; latest release v2.9.2, precompiled binaries on SourceForge are v2.8) does well. It does not do it automatically. This guide shows the manual workflows that work within the real limits of v2.9.x, so you never waste an afternoon looking for a button that does not exist.

Version note: written for v2.9.x (SourceForge binaries are 2.8). The fitting dialogs and peak functions are unchanged between the two.

Step 1 — Import the instrument export

Instrument software usually exports ASCII or CSV: wavenumber/intensity pairs, sometimes with a metadata header block. File → Import ASCII covers the common layouts; the edge cases — comment lines, transposed blocks, multi-spectrum files — are worked through in how to import data into SciDAVis. After import, assign roles with Table → Set Column As → X/Y/Z (wavenumber as X, intensity as Y), and keep one spectrum per column pair so batch comparison across samples stays sane.

Step 2 — The honest truth about peak finding and baselines

SciDAVis 2.9.x has no automatic peak finder and no one-click baseline-subtraction tool. Here is what actually works.

Marking peaks yourself. Choose Analysis → Quick Fit → Fit Multipeak. You first specify how many peaks you expect, then click each peak’s approximate position on the plot, validating each choice with the ENTER key. Those positions are only initial guesses for the fitter — the final centers, widths, and areas come from the fit. It is manual, but it is also honest: you decide how many components the chemistry justifies, and the fit cannot invent a peak you never marked.

Baselines. Every built-in peak function in SciDAVis carries a y0 offset parameter, so a flat baseline is absorbed automatically during the fit — no preprocessing needed. If the baseline slopes or curves, do not subtract it by hand first; fit it together with the peaks in the Fit Wizard (Analysis → Fit Wizard, Ctrl+Y), for example a linear term plus your peak functions. Hand-subtracting an imperfect baseline by eye and then fitting the residue is the most common source of biased peak areas in student work. Sanity check: zoom into peak-free regions — if the fitted baseline misses them visibly, the model is wrong; add a slope term or restrict the fit range before trusting any area.

The full deconvolution workflow is in multi-peak fitting for spectroscopy data.

Step 3 — Pick the right line shape

The two workhorse line shapes, exactly as defined in the official Handbook:

Rule of thumb: start with Gaussian for condensed-phase vibrational spectra and switch only if the residual pattern tells you to. And mind the A: a Gaussian A is a height while a Lorentzian A is an area — normalize to areas before any quantitative comparison across models.

For mixed models — a Gaussian shoulder on a Lorentzian main band, common in Raman/IR — Quick Fit → Fit Multipeak cannot combine families in one fit. Use the Fit Wizard with a custom sum of the two functions instead, marking each peak position as before. Judge the result by the residuals: zoom the graph and confirm they scatter randomly around zero. A systematic wiggle at a band position means a component is missing or the wrong line shape was chosen, and no amount of iteration fixes a wrong model.

Step 4 — Quantify with Beer–Lambert calibration curves

Measure your standards, plot absorbance versus concentration, and run Analysis → Quick Fit → Fit Linear. The slope is ε·l, the intercept is your blank, and the fit log gives the standard errors. Report the R² together with a residual check: a high R² with a curved residual pattern means the linear range ended earlier than you thought. Prepare the calibration in the same matrix as the samples and re-run it with every batch — the fit is only as honest as the standards behind it.

One boundary to respect: do not derive detection limits (LOD/LOQ) from the calibration fit alone. Detection limits are governed by the variability of the blank, not by how straight the standards look — run replicate blanks, summarize them with Analysis → Statistics on Columns, and combine the blank standard deviation with your calibration slope. The fit gives you the slope; the blanks give you the limit.

Notes for NMR users

Proton and carbon multiplets are sums of Lorentzian lines, so Fit Multipeak with the number of lines you expect is the natural tool; the fitted areas give relative populations. Overlapping multiplets from similar environments often need the Fit Wizard with a mixed model instead. Chemical shifts are read from the fitted xc values — more precise than eyeballing the plot cursor, and with an uncertainty from the fit.

SciDAVis for Physics & Engineering: From Sensor Data to Fitted Models

Physics and engineering run on exported data: oscilloscope traces, DAQ logs, thermocouple readings, photodiode voltages, accelerometer streams. SciDAVis — free, open-source, latest release v2.9.2 (precompiled binaries on SourceForge are v2.8) — takes those exports from raw numbers to a publication-quality plot with a fitted physical model, with no license server and no per-seat cost. This guide walks the typical physics workflow: importing and organizing sensor data, putting honest error bars on measurements, finding hidden frequency content with FFT, and extracting material parameters with nonlinear fitting.

Version note: written for v2.9.x (SourceForge binaries are 2.8). The Analysis menu, Fit Wizard dialogs, and fitting engine are unchanged between the two.

Step 1 — Import and organize your data

Most scopes and DAQs export ASCII or CSV: one column of time, one column per channel, often with a header block of instrument metadata. In SciDAVis, File → Import ASCII handles the common layouts (separator, header lines, decimal character). The tricky cases — transposed files, multi-run logs, comment lines mixed with data — are worked through step by step in how to import data into SciDAVis.

After import, tell SciDAVis which column is which: select the columns, then Table → Set Column As → X/Y/Z assigns the roles (time as X, voltage as Y; plain-text comment columns can keep describing run conditions). A little discipline here — one table per run, units in the column comments — pays off when the project grows to dozens of runs and you still need to know which trace was which.

Step 2 — Error bars that mean something

Error bars separate a student plot from a publishable one. In physics they usually come from two places: repeated measurements (the standard deviation of the repeats) or instrument specifications (a scope’s vertical accuracy, a DAQ’s least significant bit). Plot the data first, then Graph → Add Error Bars… lets you attach X errors, Y errors, or both — either as a constant value or from a column of uncertainties. Use a column whenever the uncertainty varies point to point, for example with Poisson counting statistics.

Read the result honestly: error bars are how your data tells the model “this is how far you may miss me.” A model curve that threads every error bar is suspicious (overfit); one that misses systematically is wrong (underfit). The full mechanics of the dialog — constant versus column-based errors, styling — are covered in the error bars tutorial.

Step 3 — Find hidden frequencies with FFT

Many physics signals hide periodic content: 50/60 Hz mains hum on a sensor line, a mechanical resonance in a vibration measurement, the chopper frequency in an optical experiment. Select the curve in the graph and choose Analysis → FFT. SciDAVis computes the spectrum into a new plot, and the peaks mark the characteristic frequencies of your signal. Read off the frequencies, compare with what you expect (mains, rotation rate, drive frequency), and you have either identified your noise source or confirmed your signal.

Two honest caveats. First, the frequency resolution is set by the record length, not the sampling rate — a one-second trace gives 1 Hz bins, so record longer when you need to separate close lines. Second, SciDAVis 2.9.x offers no windowing functions (no Hann or Hamming apodization), so abrupt edges of the record leak a little power into neighboring bins. For long records this is negligible; if it matters, export the CSV and apply windows in Python/SciPy. The detailed frequency-analysis workflow is in the FFT tutorial.

Step 4 — Extract parameters with nonlinear fitting

Physics rarely gives you straight lines: cooling curves are exponential, damped oscillations decay, magnetization saturates. SciDAVis offers two fitting routes, both in the Analysis menu. Quick Fit covers the standard models in one click; the Fit Wizard (Analysis → Fit Wizard, shortcut Ctrl+Y) gives full control — custom formulas, initial guesses, and fit-range limits.

A typical physics workflow: fit the model to the data, then interrogate the residuals. Zoom the graph: residuals should scatter randomly around zero. A wave, a trend, or a funnel in the residuals is the strongest signal to change the model — a high R² with a systematic residual pattern still means the wrong model. The wizard lets you restrict the x-range to the region where the model actually applies; outliers and unrelated shoulders poison the result, so fit only the physics you trust.

A concrete example: a damped oscillation from a torsion pendulum. A custom formula combining an exponential envelope with a sinusoid lets you read the damping coefficient and the natural frequency from the same fit — both parameters of the model, both with standard errors from the fit log. When Quick Fit is not enough, the custom-formula route in the wizard is where real physics fits happen. The full method, including how to read the results log and judge a fit, is in the curve fitting tutorial.

Honest limits for physics users

Two things SciDAVis will not do. It has no automatic peak finder anywhere — in multi-peak workflows you always mark the peaks yourself. And 2.9.x has no hypothesis tests (no t-test, ANOVA, or normality test); for those, export the table as CSV and use Python/SciPy or R/JASP. What it does give you — scriptable, repeatable fitting through a documented Python API — is covered in the curve fitting tutorial, and it is enough to automate a batch of runs without leaving the project.