fy gene co-expression modules in the transcriptome values of differentially expressed genes. In brief, a correlation matrix among all pairs of genes was 
created across the measured samples. Next, an adjacency matrix was constructed by raising the co-expression measure, which was a soft-threshold of the correlation matrix. The topological overlap measure was calculated based on the adjacency matrix. Then, the genes were hierarchically clustered using 1-TOM as the distance measure, and the modules were determined for the resulting dendrogram. The ME is the first principal component in a given module. The module membership of each gene in a module refers to the correlation of the gene expression level between the gene and the ME. The genes with the highest value of module membership were considered to be intramodular hub genes. The gene connection network was visualized using the bioinformatics software Cytoscape for complex networks. Custom R scripts were used to perform hierarchical clustering. To identify the differentially expressed genes in each neuron cluster, the transcriptomic data of all qualified neurons were analyzed using unsupervised hierarchical clustering and Fisher’s exact test. A negative control was obtained from clean pipettes containing bath solution. The PCR products were analyzed on ethidium bromide-stained 1.5% agarose gels. Lumbar 4 and 5 DRG were dissected from adult male mice and processed for ISH. Probes for Gal, Nppb, Il31ra, Th, Mrgpa3 and Mrgprd mRNA, among other genes, are listed in Supplementary information, In situ hybridization Immunohistochemistry Adult male mice were fixed with 4% paraformaldehyde and 1% picric acid. To label IB4-positive neurons, cryostat sections of DRG were first incubated with 5 g/ml IB4. For triple immunofluorescence, the sections were stained with goat antibodies against GSL, mouse antibodies against NF200 and rabbit antibodies against CGRP, followed by FITC-, Cy3- and Cy5-conjugated donkey secondary antibodies against goat, mouse and rabbit. The tissues were examined under a confocal R-7128 custom synthesis microscope. Single-cell PCR Sequences of the primers used in PCR are provided in Supplementary information, Mice were anesthetized with pentobarbital sodium. After laminectomy at lumbar levels 2-6, L5 DRG were exposed and isolated from the surrounding tissues. Oxygenated artificial cerebrospinal fluid was dripped periodically onto the surface of the ganglia to prevent drying and hypoxia. Two lumbar vertebrae were clamped. The epineurium covering the DRG was removed under a dissection microscope. The skin was sewn to a ring to maintain a pool of warm oxygenated ACSF. The DRG was continuously superfused at a rate of 3 ml/min by oxygenated ACSF that was preheated by an in-line heater to achieve the desired temperature of 37 C. The ACSF contained 130 NaCl, 3.5 KCl, 24 NaHCO3, 1.25 NaH2PO4, 1.2 MgCl2, 1.2 CaCl2 and 10 dextrose. The solution was bubbled with 95% O2 and 5% CO2, and had a pH of 7.4 and an osmolarity of 290-310 Osm. The vasoconstricting drug -vasopressin was applied via a syringe. The va- In vivo whole-cell patch-clamp recording and neuron classification npg Types of primary sensory neurons 100 soconstrictor stopped or substantially reduced the blood flow to the capillary bed in the DRG, thereby preventing bleeding that PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19822652 would otherwise occur after the application of collagenase. Collagenase P was applied to a local region of the DRG via a pipette located close to the surface. The bath perfusio