Alkan_SciSignal2018 estimates noise parameters even thought they are fixed in data2dynamics

[FFroehlich]

data2dynamics reports qFit=0 for all standard deviations (sd_count_abs, sd_count_casp_abs, sd_patm_au, sd_pdnapk_au, sd_pchk1_au, sd_pchk2_au, sd_tp53_au, sd_pp53_au, sd_tp21_au, sd_yh2ax_au, sd_p21_mRNA_fold, sd_wip1_mRNA_fold) and some of the noise parameters have suspiciously round values (sd_count_abs = sd_count_casp_abs = sd_pchk1_au = 0.1, sd_p21_mRNA_fold = sd_wip1_mRNA_fold = 1.0) or consistent values (sd_pdnapk_au = sd_pp53_au = sd_tp53_au = 0.19952623149688797) making it hard to believe they were ever estimated.

[FFroehlich]

I can now also confirm that fixing those standard deviations to be non-estimated reproduces the nominal optimum after gaussian noise correction up to numerical precision.