More Spike Sorting

1
More Spike Sorting

• Kenneth D. Harris
• Rutgers University

2
How Do You Know It Works?

• We can split waveforms into clusters, but are we
  sure they correspond to single cells?
• Simultaneous intra- and extra-cellular recordings
  allow us to estimate errors.
• Quality measures allow us to guess errors even
  without simultaneous intracellular recording.

3
Intra-extra Recording

• Simultaneous recording with a wire tetrode and
  glass micropipette.

4
Intra-extra Recording
Extracellular waveform is almost minus derivative
of intracellular
5
Bizarre Extracellular Waveshapes
Model
Experiment
6
Waveshape Helps Separation
7
Two Types of Error

• Type I error (false positive)
• Incorrect inclusion of noise, or spikes of other
  cells
• Type II error (false negative)
• Omission of true spikes from cluster
• Which is worse? Depends on application

8
Manual Clustering Contest
9
Best Ellipsoid Error Rates
Find ellipsoid that minimizes weighted sum of
Type I and Type II errors. Must evaluate using
cross-validation!
10
Humans vs. B.E.E.R.
11
Why were human errors so high?

• To understand this, try to understand why
  clusters have the shape they do
• Simplest possibility spike waveform is constant,
  cluster spread comes from background noise
• Clusters should be multivariate normal

12
Problem Overlapping Spikes
13
Problem Cellular Synchrony
14
Problem Bursting
15
Problem Misalignment

• When you have a spike whose peak occurs at
  different times on different channels, it can
  align on either.
• This causes the cluster to be split in two.

16
Problem Dimensionality
Manual clustering only uses 2 dimensions at a
time BEER measure can use all of them
17
Automatic Clustering

• Uses all dimensions at once
• Errors should be lower
• Still requires human input

18
Human-machine Interface
19
Semi-automatic Performance
20
Cluster Quality Measures

• Would like to automatically detect which cells
  are well isolated.
• BEER measure needs intracellular data, which we
  dont have in general.
• Will define two measures that only use
  extracellular data.

21
Isolation Distance
Size of ellipsoid within which as many spikes
belong to our cluster as not
22
L_ratio
23
False Positives and Negatives
24
Which Measure to Use?

• Isolation distance correlates with false positive
  error rates
• Measures distance to other clusters
• L_ratio correlates with false negative error
  rates
• Measures number of spikes near cluster boundary

25
Conclusions

• Automatic clustering will save time and reduce
  errors.
• Errors can be as low as 5.
• Quality measures give you a feeling of how bad
  your errors are.

26
Room for Improvement
Easy

• Make it faster
• Better human-machine interaction
• Improved spike detection and alignment
• Quality measures that estimate error
• Fully automatic sorting
• Resolve overlapping spikes

Hard