Toshiya Arakawa1, Akira Tanave2, Shiho Ikeuchi3, Aki Takahashi4, Satoshi Kakihara5, Shingo Kimura6, Hiroki Sugimoto7, Nobuhiko Asada3, Toshihiko Shiroishi8, Kazuya Tomihara6, Takashi Tsuchiya9, Tsuyoshi Koide10. 1. Department of Mechanical System Engineering, Aichi University of Technology, Gamagori, Aichi 443-0047, Japan. 2. Mouse Genomics Resource Laboratory, National Institute of Genetics (NIG), Mishima, Shizuoka 411-8540, Japan; Department of Genetics, SOKENDAI, Mishima, Shizuoka 411-8540, Japan. 3. Okayama University of Science, Okayama 700-0005, Japan. 4. Mouse Genomics Resource Laboratory, National Institute of Genetics (NIG), Mishima, Shizuoka 411-8540, Japan; Department of Genetics, SOKENDAI, Mishima, Shizuoka 411-8540, Japan; Transdisciplinary Research Integration Center, Research Organization of Information and Systems, Minatoku, Tokyo 105-0001, Japan. 5. Graduate School of Information Science and Technology, The University of Tokyo, Bunkyo-ku, Tokyo 113-8656, Japan. 6. Department of Psychology, Faculty of Law, Economics and Humanities, Kagoshima University, Kohrimoto, Kagoshima 890-0065, Japan. 7. Mouse Genomics Resource Laboratory, National Institute of Genetics (NIG), Mishima, Shizuoka 411-8540, Japan; Transdisciplinary Research Integration Center, Research Organization of Information and Systems, Minatoku, Tokyo 105-0001, Japan. 8. Department of Genetics, SOKENDAI, Mishima, Shizuoka 411-8540, Japan; Transdisciplinary Research Integration Center, Research Organization of Information and Systems, Minatoku, Tokyo 105-0001, Japan; Mammalian Genetics Laboratory, NIG, Mishima, Shizuoka 411-8540, Japan. 9. National Graduate Institute for Policy Studies, Minato-ku, Tokyo 106-8677, Japan; The Institute of Statistical Mathematics, Tachikawa, Tokyo 190-8562, Japan. 10. Mouse Genomics Resource Laboratory, National Institute of Genetics (NIG), Mishima, Shizuoka 411-8540, Japan; Department of Genetics, SOKENDAI, Mishima, Shizuoka 411-8540, Japan; Transdisciplinary Research Integration Center, Research Organization of Information and Systems, Minatoku, Tokyo 105-0001, Japan. Electronic address: tkoide@nig.ac.jp.
Abstract
BACKGROUND: Owing to their complex nature, social interaction tests normally require the observation of video data by a human researcher, and thus are difficult to use in large-scale studies. We previously established a statistical method, a hidden Markov model (HMM), which enables the differentiation of two social states ("interaction" and "indifference"), and three social states ("sniffing", "following", and "indifference"), automatically in silico. NEW METHOD: Here, we developed freeware called DuoMouse for the rapid evaluation of social interaction behavior. This software incorporates five steps: (1) settings, (2) video recording, (3) tracking from the video data, (4) HMM analysis, and (5) visualization of the results. RESULTS: Using DuoMouse, we mapped a genetic locus related to social interaction. We previously reported that a consomic strain, B6-Chr6C(MSM), with its chromosome 6 substituted for one from MSM/Ms, showed more social interaction than C57BL/6 (B6). We made four subconsomic strains, C3, C5, C6, and C7, each of which has a shorter segment of chromosome 6 derived from B6-Chr6C, and conducted social interaction tests on these strains. DuoMouse indicated that C6, but not C3, C5, and C7, showed higher interaction, sniffing, and following than B6, specifically in males. COMPARISON WITH EXISTING METHOD: The data obtained by human observation showed high concordance to those from DuoMouse. The results indicated that the MSM-derived chromosomal region present in C6-but not in C3, C5, and C7-associated with increased social behavior. CONCLUSIONS: This method to analyze social interaction will aid primary screening for difference in social behavior in mice.
BACKGROUND: Owing to their complex nature, social interaction tests normally require the observation of video data by a human researcher, and thus are difficult to use in large-scale studies. We previously established a statistical method, a hidden Markov model (HMM), which enables the differentiation of two social states ("interaction" and "indifference"), and three social states ("sniffing", "following", and "indifference"), automatically in silico. NEW METHOD: Here, we developed freeware called DuoMouse for the rapid evaluation of social interaction behavior. This software incorporates five steps: (1) settings, (2) video recording, (3) tracking from the video data, (4) HMM analysis, and (5) visualization of the results. RESULTS: Using DuoMouse, we mapped a genetic locus related to social interaction. We previously reported that a consomic strain, B6-Chr6C(MSM), with its chromosome 6 substituted for one from MSM/Ms, showed more social interaction than C57BL/6 (B6). We made four subconsomic strains, C3, C5, C6, and C7, each of which has a shorter segment of chromosome 6 derived from B6-Chr6C, and conducted social interaction tests on these strains. DuoMouse indicated that C6, but not C3, C5, and C7, showed higher interaction, sniffing, and following than B6, specifically in males. COMPARISON WITH EXISTING METHOD: The data obtained by human observation showed high concordance to those from DuoMouse. The results indicated that the MSM-derived chromosomal region present in C6-but not in C3, C5, and C7-associated with increased social behavior. CONCLUSIONS: This method to analyze social interaction will aid primary screening for difference in social behavior in mice.