✎ analysis:nsb []

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
Welcome! This is the home page for the data management and analysis tutorials for the NS&B 2022 mouse module.

=== Contents ===

== Reference: skim these so you know what's here ==

   * [[analysis:nsb2015:week0|Principles of (neural) data analysis]]
   * [[analysis:matlab-overview|Overview of common functions]]
   * [[analysis:matlab-examples|Example data analysis workflows]]

== Fundamentals: do these in Week 1 ==

   * [[analysis:nsb2019:week1|Module 1: Setting up (MATLAB, paths, GitHub, accessing data)]]
   * [[analysis:nsb2017:week2|Module 2: Introduction to neural data formats and preprocessing]]
   * [[analysis:nsb2016:week3long|Module 3: Visualizing raw neural data in MATLAB]] ([[analysis:nsb2016:week3short|Short version]])
   * [[analysis:nsb2017:spikesort|Module 4: Spike sorting tetrode data]]
 
== Time series data data basics: do as needed ==

   * [[analysis:nsb2016:week4|Module 5: Anatomy of time series data, sampling theory]]
   * [[analysis:nsb2016:week5|Module 6: Fourier series, transforms, power spectra]]
   * [[analysis:nsb2016:week6|Module 7: Filtering: filter design, use, caveats]]
   * [[analysis:nsb2016:week7|Module 8: Time-frequency analysis: spectrograms]]

== Spike data basics: do as needed ==

   * [[analysis:nsb2016:week9|Module 9: Spike train analysis: firing rate, interspike interval distributions, auto- and crosscorrelations]]
   * [[analysis:nsb2016:week10|Module 10: Spike train analysis II: tuning curves, encoding, decoding]]


== Intermediate topics: do as needed ==

   * [[analysis:nsb2016:week11|Module 11: Interactions between multiple signals: coherence, Granger causality, and phase-slope index]]
   * [[analysis:nsb2016:week12|Module 12: Time-frequency analysis II: cross-frequency coupling]] 
   * [[analysis:nsb2016:week13|Module 13: Spike-field relationships: spike-triggered average, phase locking, phase precession]]
   * [[analysis:nsb2016:week14|Module 14: Classification of ensemble spiking patterns]]

== Advanced topics: do as needed ==

  * [[analysis:nsb2016:week15|Module 15: Two-step Bayesian decoding with dynamic spatial priors]]
  * [[analysis:nsb2016:week16|Module 16: Pairwise co-occurrence (replay)]]

== Other topics ==

  * Git: conflict resolution, undo's, writing good commit messages, issue tracking, branching (on request)
  * Top-level analysis workflows for handling multiple subjects and sessions (on request)
  * Exporting MATLAB data to R (on request)
  * MATLAB tools: GUI design tool, debugger, profiler (on request)

=== Prerequisites ===

Basic familiarity with MATLAB. Depending on your background and programming experience you might find the following resources helpful:

  * Textbook: {{:analysis:wallisch_matlabforneuro.pdf|Wallisch, MATLAB for Neuroscientists}}
  * [[http://www.mathworks.com/help/matlab/getting-started-with-matlab.html?s_cid=learn_doc|"Getting Started with MATLAB" Primer]]. 
  * [[http://www.mathworks.com/matlabcentral/about/cody/ | Cody]], a continually expanding set of problems with solutions to work through, with a points system to track your progress

If you are unsure, take a look at the table of contents of the Primer. If there are things you don't recognize, use the Primer itself, or Chapter 2 of the MATLAB for Neuroscientists book to get up to speed.

Regardless of your MATLAB abilities, some great ways to keep learning are:

  * [[http://blogs.mathworks.com/ | Mathworks staff blogs]], especially "Loren on the Art of MATLAB" is a treasure trove of tips and tricks
  * [[http://stackoverflow.com/questions/tagged/matlab | MATLAB questions on StackOverflow]], a Q&A site where you can browse previous questions and add new ones

If you have no formal training in computer programming (i.e. you have never taken a "Intro to Computer Science" or "Introductory Programming" type course) you will almost certainly find what follows in this course less frustrating if you do the pen-and-paper exercises in this [[http://sites.tufts.edu/rodrego/files/2011/03/Secrets-of-Computer-Power-Revealed-2008.pdf | short chapter]] by Daniel Dennett ("The Secrets of Computer Power Revealed") before you embark on the MATLAB primer linked to above.

=== Resources ===

This course is "standalone", but the following textbooks provide more in-depth treatment of some of the topics.

  * Textbook: {{:analysis:leis_dspusingmatlab.pdf|Leis, Digital Signal Processing using MATLAB for Students and Researchers}}
  * Textbook: {{:analysis:johnstonwu.pdf|Johnston and Wu, Foundations of Cellular Neurophysiology}}
  * Textbook: {{:analysis:dayanabbott_theoneuro.pdf|Dayan & Abbott, Theoretical Neuroscience}}

=== What this tutorial provides ===

Overall, this tutorial is designed to provide hands-on experience with management, visualization, and analysis of neural data. Becoming skilled at these things is a rate-limiting step for many graduate projects requiring analysis. Even if your work only requires rudimentary analysis, awareness of what else can be done and how to do it well is valuable, for instance when evaluating the work of others in the literature!

To do so, the focus is on introducing some commonly used tools, such as %%GitHub%% and relevant functionality within MATLAB -- and then to actually use these on real data sets you collect yourself.

We will make contact with a few concepts from computer science, signal processing, and statistics. However, the focus is on making initial steps that work and getting pointers to more complete treatment, rather than a thorough theoretical grounding. Nevertheless, to make sure that what you learn is not tied to specific data sets only, a number of [[analysis:nsb2015:week0|principles]] of data analysis -- applicable to any project of sufficient complexity -- will be referenced throughout the material. You are invited to think of these and others, not only as you progress through the modules, but especially as you organize your own data analyses!

=== What this tutorial is not ===

This material will provide a brief introduction to a number of concepts which are themselves the subject of multiple courses and voluminous textbooks. These include signal processing topics such as Fourier analysis and filter design, computer science concepts such as object-oriented programming and binary data formats, and a number of statistical ideas and tools. Be aware that if any of these are particularly important to your research, you should consider taking more in-depth coursework and/or working through relevant textbooks on your own: this short tutorial cannot replace such courses!

=== Acknowledgments ===

The architecture of the code used here was inspired by a similar set of code by my post-doctoral mentor, [[http://redishlab.neuroscience.umn.edu/ | A. David Redish]]; several of the data types and functions are re-implementations of Redish lab functions of the same name. Major contributions to the codebase were made by Alyssa Carey (a former grad student in the lab and NS&B TA) and Youki Tanaka (former grad student and NS&B alum).
User Tools

Site Tools

Page Tools
