Inference of monosynaptic connections from parallel spike trains: A review

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Neuronal connectivity is crucial for understanding brain function.
• Different types of connectivity exist: structural, functional, and monosynaptic.
• Inferring direct neuronal links from neural data is a key challenge.

Purpose of the Study:

• To review progress in inferring monosynaptic connectivity from spike train data over the last 20 years.
• To categorize and explain major inference methodologies.
• To discuss current tools and future research directions.

Main Methods:

• Focus on methods for inferring monosynaptic connectivity from multi-neuron spike trains.
• Summarize correlation-based approaches.
• Summarize model-based approaches.

Main Results:

• Significant advancements in inferring monosynaptic connections from spike data have been achieved.
• Both correlation-based and model-based methods have contributed to improved accuracy.
• Available source codes for connectivity inference are discussed.

Conclusions:

• The accuracy of inferring monosynaptic connections has dramatically improved.
• Despite ongoing challenges, current methods offer powerful tools for network analysis.
• Future efforts will continue to refine these inference techniques.