− | sloppiness是多参数系统的中常见的一种特性。具有这种特性的模型的参数往往有很多个,但是模型的行为仅取决于少数几个参数或参数的线性组合,其它参数或参数 的线性组合对模型的影响微乎其微。sloppiness特性在系统生物学<ref>Panas D, Amin H, Maccione A, Muthmann O, van Rossum M, Berdondini L, Hennig MH (2015) 1. Sloppiness in spontaneously active neuronal networks.,Panas D, Amin H, Maccione A, Muthmann O, van Rossum M, Berdondini L, Hennig MH,J Neurosci. 2015 Jun 3;35(22):8480-92. doi: 10.1523/JNEUROSCI.4421-14.2015. PMID: 26041916; PMCID: PMC4452554</ref>、物理学和数学系统中无处不在。 | + | sloppiness是多参数系统的中常见的一种特性。具有这种特性的模型的参数往往有很多个,但是模型的行为仅取决于少数几个参数或参数的线性组合,其它参数或参数 的线性组合对模型的影响微乎其微。sloppiness特性在系统生物学<ref>Sloppiness in spontaneously active neuronal networks.,Panas D, Amin H, Maccione A, Muthmann O, van Rossum M, Berdondini L, Hennig MH,J Neurosci. 2015 Jun 3;35(22):8480-92. doi: 10.1523/JNEUROSCI.4421-14.2015. PMID: 26041916; PMCID: PMC4452554</ref><ref>Cortical state transitions and stimulus response evolve along stiff and sloppy parameter dimensions, respectively.Adrian Ponce-Alvarez,Gabriela Mochol, Ainhoa Hermoso-Mendizabal,Jaime de la Rocha ,Gustavo Deco, (2020)eLife 9:e53268.</ref><ref>Sloppy models and parameter indeterminacy in systems biology: [https://arxiv.org/abs/q-bio/0701039 "Universally Sloppy Parameter Sensitivities in Systems Biology"], Ryan N. Gutenkunst, Joshua J. Waterfall, Fergal P. Casey, Kevin S. Brown, Christopher R. Myers, James P. Sethna, PLoS Comput Biol3(10) e189 (2007). ([http://compbiol.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pcbi.0030189 PLoS], [https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.0030189 doi:10.1371/journal.pcbi.0030189]), [https://sethna.lassp.cornell.edu/pubPDF/SloppyEverywhere.pdf pdf]). [Reviewed in [http://biomedicalcomputationreview.org/4/1/4.pdf NewsBytes] of [http://biomedicalcomputationreview.org/ Biomedical Computation Review] (Winter 07/08); rated "Exceptional" on Faculty of 1000]. </ref>、物理学<ref>Sloppiness, information geometry, and model reduction: [https://arxiv.org/abs/1501.07668 Perspective: Sloppiness and Emergent Theories in Physics, Biology, and Beyond], Mark K. Transtrum, Benjamin B. Machta, Kevin S. Brown, Bryan C. Daniels, Christopher R. Myers, and James P. Sethna, [https://pubs.aip.org/aip/jcp/article/143/1/010901/566995/Perspective-Sloppiness-and-emergent-theories-in J. Chem. Phys. '''143''', 010901 (2015)], </ref>和数学<ref>[https://gutengroup.mcb.arizona.edu/wp-content/uploads/Mannakee2016.pdf Sloppiness, information geometry, and model reduction: Sloppiness and the geometry of parameter space,] Mannakee B.K., Ragsdale A.P., Transtrum M.K., Gutenkunst R.N., [https://link.springer.com/chapter/10.1007/978-3-319-21296-8_11 Uncertainty in Biology, Volume 17 of the series Studies in Mechanobiology, Tissue Engineering and Biomaterials], </ref>系统中无处不在。 |
− | 几年前,在研究细胞内外信号传递过程中蛋白质相互作用机制时,几名物理和生物领域的科学家建立了一个有48个参数的模型,模型中参数之间难以独立分离,且参数变化范围都超过50倍。在面对一个参数不确定性如此之大的复杂模型时,一位生物学家却指出:根据研究经验,模型的实验结果甚至不用电脑就可以估算出来。因为系统的行为与大多数参数不确定性之间的关系并不紧密。
| + | 几年前,在研究细胞内外信号传递过程中蛋白质相互作用机制<ref>[https://sethna.lassp.cornell.edu/pubPDF/PC12.pdf The Statistical Mechanics of Complex Signaling Networks: Nerve Growth Factor Signaling], Kevin S. Brown, Colin C. Hill, Guillermo A. Calero, Christopher R. Myers, Kelvin H. Lee, James P. Sethna, and Richard A. Cerione, Physical Biology 1, 184-195 (2004), with [https://sethna.lassp.cornell.edu/pubPDF/PC12Supporting.pdf supplemental material]. </ref><ref>[https://sethna.lassp.cornell.edu/pubPDF/SloppyModelPRE.pdf "Statistical Mechanics Approaches to Models with Many Poorly Known Parameters"], Kevin S. Brown and James P. Sethna, Phys. Rev. E 68, 021904 (2003). </ref>时,几名物理和生物领域的科学家建立了一个有48个参数的模型,模型中参数之间难以独立分离,且参数变化范围都超过50倍。在面对一个参数不确定性如此之大的复杂模型时,一位生物学家却指出:根据研究经验,模型的实验结果甚至不用电脑就可以估算出来。因为系统的行为与大多数参数不确定性之间的关系并不紧密。 |
− | 即便参数值与真实值相差很大,有sloppy特性的模型也可以做出精确的预测。在数学中有一个经典的拟合难题:用指数衰变和去拟合放射性模型(c列和d列)得到的衰变常数与真实衰变常数截然不同,但短期内模型预测值与真实值却相差不大 。最后,用多项式系数模型<math>\sum_i a_it^i</math>拟合数据是sloppy的(h列)。但用正交多项式基<math>\sum_ib_iH_i</math>(<math>H_i</math>是一组正交多项式基)去拟合时得到的模型却往往是非sloppy的,这是因为从<math>t^i</math>到<math>H_i</math>的变换是高度非正交的。
| + | 即便参数值与真实值相差很大,有sloppy特性的模型也可以做出精确的预测。在数学中有一个经典的拟合难题<ref>Formulation, application to fitting algorithms: [https://arxiv.org/abs/0909.3884 "Why are nonlinear fits to data so challenging?"], Mark K. Transtrum, Benjamin B. Machta, and James P. Sethna, [https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.104.060201 Phys. Rev. Lett.] '''104''', 060201 (2010).</ref>:用指数衰变和去拟合放射性模型(c列和d列)得到的衰变常数与真实衰变常数截然不同,但短期内模型预测值与真实值却相差不大 。最后,用多项式系数模型<math>\sum_i a_it^i</math>拟合数据是sloppy的(h列)。但用正交多项式基<math>\sum_ib_iH_i</math>(<math>H_i</math>是一组正交多项式基)去拟合时得到的模型却往往是非sloppy的,这是因为从<math>t^i</math>到<math>H_i</math>的变换是高度非正交的。 |