A course every science undergraduate should take?

Science is becoming increasingly inter-disciplinary.
Most science graduates do not end up working in scientific research or an area related to their undergraduate major.
Yet most undergraduate curricula are essentially the same as they were fifty years ago and are copies of courses at MIT-Berkeley-Oxford designed for people who would go on to a Ph.D and (hopefully) end up working in research universities.
Biology and medicine are changing rapidly particularly in becoming more quantitative.
How should we adapt to these realities?
Are there any existing courses that might be appropriate for every science major to take?

Sometimes my colleagues get upset that advanced physics undergrads don't know certain things they should [Lorentz transformations, Brownian motion, scattering theory, ....]. But, my biggest concern is that they don't have certain basic skills [dimensional analysis, sketching graphs, recognising silly answers,  writing clearly, ....]. These skills will be important in almost any job that has a quantitative dimension to it.

For the past few years Phil Nelson has been teaching a course at University of Pennsylvania that I think may "fit the bill." The associated textbook Physical Models of Living Systems will be published at the end of the year.

Previously, I have lavished praise on Nelson's book Biological Physics: Energy, Information, Life. I used it in a third year undergraduate biophysics course PHYS3170 and think it is one of the best textbooks I have every encountered. Besides clarity and fascinating subject material it has excellent problems [and solutions manual], makes use of real experimental data, has informative section headings, often discusses the limitations of different approaches, and uses nice historical examples.

The new book covers different material and focuses on some particular skills that any science and engineering major should learn, regardless of whether they end up working in biology or medicine or biophysics. Below I reproduce some of Nelson's summary.

Readers will acquire several research skills that are often not addressed in traditional courses:

• Basic modeling skills, including dimensional analysis, identification of variables, and ODE formulation.
• Probabilistic modeling skills, including stochastic simulation.
• Data analysis methods, including maximum likelihood and Bayesian methods.
• Computer programming using a general-purpose platform like MATLAB or Python, with short codes written from scratch.
• Dynamical systems, particularly feedback control, with phase portrait methods.

[Here modeling does not mean running pre-packaged computer software but developing simple physical models].

All of these basic skills, which are relevant to nearly any field of science or engineering, are presented in the context of case studies from living systems, including:

• Virus dynamics
• Bacterial genetics and evolution of drug resistance
• Statistical inference
• Superresolution microscopy
• Synthetic biology
• Naturally evolved cellular circuits, including homeostasis, genetic switches, and the mitotic clock.

This looks both important and fascinating. The Instructors Preface makes an excellent case for the importance of the course, including to pre-medical students. The Table of Contents illustrates not just the logical flow and interesting content but again uses informative section headings that summarise the main point.

So, what do you think?

Is this a course (almost) every science undergraduate should take?

Are there specific courses you think all students should take?