Spring 2020, 9:30-10:45 TTh 221 MLH (MacLean Hall)

**Instructors:**

Sriram V. Pemmaraju

101G MLH, sriram-pemmaraju@uiowa.edu, 319-353-2956

Office Hours: 1:30-2:30 M, 10:30-11:30 W, 2:00-3:00 F.

Alberto M. Segre

14G MLH, alberto-segre@uiowa.edu, 319-335-1713

Office: TBA.

Our office hours are "walk-in" hours and you don't need to make an appointment to see us during office hours. We are also happy to meet with students outside office hours by prior appointment.

**Course webpage:** `homepage.cs.uiowa.edu/~sriram/4980/spring20/`

**Department website:** `http://www.cs.uiowa.edu/`

This is a graduate-level Computer Science (CS) course on *computational epidemiology*, which
is the study and development of computational techniques and tools for modeling, simulating, predicting, forecasting,
surveilling, mitigating, and visualizing the spread of disease.
In this course, we will use techniques from different areas of CS including algorithms, data mining,
discrete-event simulations, machine learning, and network science.
The course is organized into four parts: (i) Disease-spread models and analysis of disease dynamics,
(ii) Inference, prediction, and forecasting problems related to disease-spread, (iii) Infection control and
disease surveillance problems, (iv) Additional topics including a discussion of disease-related datasets and
the use of technology for gathering contact data.
A more detailed list of topics and readings appears further below.

No prior background in epidemiology or biology is assumed. However, a solid background in discrete mathematics, especially graph theory and discrete probability, and a solid background in programming and data structures will be assumed. Background in linear algebra and statistics is not required, but mathematical maturity in these areas will likely be helpful. A substantial portion of student evaluation will be via a group project, that will include an end-of-term technical paper and presentation. Additional modes of evaluation will include solving homework problems, writing short technical pieces, participating in classroom and online discussions, and scribing lecture notes. There will be no exams in this course.

Syllabus document, Announcements, Assignments, Weekly Topics

- For your use, sample latex source files for scribe notes are available in this directory.
- Updates on the 2019 coronavirus (COVID-19), from UI and from the CDC.
- COVID-19 resources for projects:
- COVID tracking project
- COVID-19 reports from the Imperial College, London
- COVID-19 social distancing scoreboard
- Slate article on social distancing as measured by NYC metro turnstile deta
- R package for COVID-19 modeling and visualization, with links to mostly Chinese data sources

- Reading Response 1. Due in class Tue, 1/28.
- Reading Response 2. Due in class Thu, 2/20.
- Project Plans.
- Links to external datasets and presentation describing the UIHC datasets.
- Reading Response 3. Due before class Thu, 4/2.

**1/20-1/24**- Introduction. What is epidemiology? What is computational epidemiology? Brief history of epidemiology and computational epidemiology. Bernoulli's math model for smallpox from 1760s. John Snow and the Broad Street Cholera outbreak. Main themes of the course. slides
- Recent research on C.diff infections (CDI) at the University of Iowa Hospitals and Clinics (UIHC). slides

**Readings**:- The next plague is coming: Is America ready?
by Ed Yong, The Atlantic, July/Aug 2018.

**1/27-1/31**- Recent research on statistical tests that show spatio-temporal clustering of C.diff infections (CDI) at the University of Iowa Hospitals and Clinics (UIHC). slides
- Introduction to
*compartmental*models. Analysis of the compartmental SIR model. Other related models such as SIS and SEIR. Definition of the*basic reproductive number*and its implications to disease-spread.

**Readings**:- Writeup on the Knox test.
- SpatioTemporal Clustering of in-Hospital Clostridioides difficile Infection (CDI), Pai et al. Draft of paper to appear in ICHE 2020.
- The next plague is coming: Is America ready?
by Ed Yong, The Atlantic, July/Aug 2018.

- Sections 1 and 2 from Mathematical Approaches to Infectious Disease
Prediction and Control, Dimitrov and Meyers. INFORMS: Tutorials in Operations Research, 2010.

- Novel coronavirus 2019-nCoV: early estimation of epidemiological parameters and epidemic predictions, 27 Jan 2020, Read et al.

**2/3-2/7**- Variants of the SIR model: SEIR, SIS, MSIR, SIRS, etc., limitations of compartmental models and motivation for contact network modeling of disease spread,
the
*cascades*model for disease-spread on contact networks, the definition and role of R0, connections to percolation, implications for computation.

**Readings**:- Sections 4 and 5.1 (including the introductory portion of Section 5) from Mathematical Approaches to Infectious Disease
Prediction and Control, Dimitrov and Meyers. INFORMS: Tutorials in Operations Research, 2010.

- Sections 21.1, 21.2, and 21.3 from Networks, Crowds, and Markets: Reasoning about a Highly Connected World by Easley and Kleinberg.

- Variants of the SIR model: SEIR, SIS, MSIR, SIRS, etc., limitations of compartmental models and motivation for contact network modeling of disease spread,
the
**2/10-2/14**- Case study of a contact network of healthcare workers from the University of Iowa Hospitals and Clinics (UIHC), generated from data.
- Features of contact network that are typically modeled. Degree distributions and
*configuration model*graphs.*Power-law*degree distributions. -
*Small-world*networks and implications for disease-spread.

**Readings**:- Healthcare Worker Contact Networks and the Prevention of Hospital-Acquired Infections, Curtis et al., PLOS One, December 2013.
- Collective dynamics of "small-world" networks, Watts and Strogatz, Nature, June 1998.
- Nowcasting and forecasting the potential domestic and international spread of the 2019-nCoV outbreak originating in Wuhan, China: a modelling study, Wu, Leung, and Leung, The Lancet, Jan 31 2020.

**2/24-2/28**- Introduction to
*inference problems*for disease models, Markov chain Monte Carlo (MCMC) methods for inferring disease parameters in compartmental models. - Inferring infection sources in contact network models.

**Readings**:- Model fitting and inference for infectious disease dynamics, by Funk et al. We will focus on the first two sections: "Introduction" and "MCMC." Most of the material covered in class will be from the lecture slides for these sections.
- Detecting sources of computer viruses in networks: theory and experiment, by Shah and Zaman, SIGMETRICS 2010.

- Introduction to
**3/9-3/13**- Inferring the role of in-hospital spread of C.diff infection using simple logistic regression models.
- Inferring the role of asymptomatic spreaders in contact network models.

**Readings**:- Evaluation of Clostridium difficile-Associated Disease Pressure as a Risk Factor for C difficile Associated Disease, by Dubberke et al., Archives of Internal Medicine, 2007.
- Learning the Probability of Activation in the Presence of Latent Spreaders, by Makar, Guttag and Wiens, AAAI 2018.

**3/30-4/3**- Introduction to the design and evaluation of disease control policies.
- Report on the impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality and healthcare demand. This is a report published on March 16th by modelers and epidemiologists from the Imperial College, London.
- An algorithmic formulation of the vaccine allocation problem.

**Slides and videos**:**Tue 3/31**slides, video, and discussion questions.

**Readings**:- Start reading from (and including) the subsection "Design of effective vaccination policies" from Healthcare Worker Contact Networks and the Prevention of Hospital-Acquired Infections, Curtis et al., PLOS One, December 2013. Focus on Figures 4, 5, and 6.
- Section 6 from Mathematical Approaches to Infectious Disease
Prediction and Control, Dimitrov and Meyers. INFORMS: Tutorials in Operations Research, 2010.

- Report 9: Impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality and healthcare demand by Neil M. Ferguson et al, March 16th 2020.
- Unbalanced Graph Cuts, by Hayrapetyan, Kempe, Pal, and Svitkina ESA 2005.