[us-commits] [ehb54/ultrascan3] 2d5bca: Normalize by concentration before per-q regression...

[emre brookes]

  Branch: refs/heads/somo-dev-claude
  Home:   https://github.com/ehb54/ultrascan3
  Commit: 2d5bca3b5b3a7110db6eefcdb28ef3ff644fad9c
      https://github.com/ehb54/ultrascan3/commit/2d5bca3b5b3a7110db6eefcdb28ef3ff644fad9c
  Author: ehb54 <brookes at uthscsa.edu>
  Date:   2026-06-21 (Sun, 21 Jun 2026)

  Changed paths:
    M us_somo/develop/src/us_hydrodyn_saxs_iqq_extrap_c0.cpp

  Log Message:
  -----------
  Normalize by concentration before per-q regression in do_extrap_c0

Raw SAXS intensity is dominated by a term directly proportional to
concentration, so fitting raw I(q) vs c trivially extrapolates to ~0
at c=0 -- confirmed on real asyn data, where the output curve was
just noise oscillating around zero. The standard SAXS technique (the
Zimm-plot analogue) fits I(q,c)/c against c instead; its c=0 intercept
recovers the ideal, structure-factor-free dilute-limit curve.

Concentration (x-axis) stays the real, distinct entered values --
only the intensity (y-axis) is divided by each curve's own
concentration. Curves with concentration <= 0 are excluded from the
regression (can't normalize by zero) with a warning. Output curve is
now in concentration-normalized units, noted in the summary message.

Validated against the 6 real asyn .dat files: the normalized
intercept now produces a smooth, positive, Guinier-shaped curve
instead of near-zero noise.

Co-Authored-By: Claude Sonnet 4.6 <noreply at anthropic.com>



To unsubscribe from these emails, change your notification settings at https://github.com/ehb54/ultrascan3/settings/notifications