BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Talks.cam//talks.cam.ac.uk// X-WR-CALNAME:Talks.cam BEGIN:VEVENT SUMMARY:CNN Seminar - November - Sarah Teichman (MRC Laboratory of Molecul ar Biology) and Daniele Quercia (Horizon Researcher\, Computer Labs) DTSTART:20111121T143000Z DTEND:20111121T153000Z UID:TALK34604@talks.cam.ac.uk CONTACT:Petra Vertes DESCRIPTION:Abstracts:\n\n1) Network Analyses of Protein Structure\, Evolu tion and Dynamics\n\nGraph theory has made a huge impact on molecular biol ogy in the past ten years as a paradigm for describing gene and protein in teractions. In our group\, we have used graph theoretical approaches to an alyse the topology and evolution of genome-scale networks of both physical protein interactions [1] and transcriptional regulatory interactions [2]. This has shed light on the pervasive role of gene duplication in shaping both types of networks. We have also used networks to analyse dynamical p rocesses in protein protein [3] and transcriptional regulatory interaction s [4]. More recently\, we have introduced the concept of networks to descr ibe interactions between proteins within small complexes of known three-di mensional structure [5]. This has proved tremendously powerful in elucidat ing the principles of protein assembly and evolution.\n\n[1] Pereira-Leal\ , J. & Teichmann\, S.A. (2005) Novel specificities emerge by step-wise dup lication of functional modules. Genome Res.\, 15\, 552-559.\n\n[2] Teichma nn\, S.A. & Babu\, M.M. (2004) Gene Regulatory Network Growth by Duplicati on. Nature Genet.\, 36\, 492-496.\n\n[3] Levy\, E.D.\, Boeri-Erba\, E.\, R obinson\, C.V. & Teichmann\, S.A. (2008) Assembly reflects evolution of pr otein complexes. Nature\, 453\, 1262-5\n\n[4] Luscombe\, N.M.\, Babu\, M.M .\, Yu\, H.\, Snyder\, M.\, Teichmann\, S.A. & Gerstein\, M. (2004) Genome -scale analysis of regulatory network dynamics. Nature\, 431\, 308-312.\n\ n[5] Levy\, E.\,D. Pereira-Leal\, J.\, Chothia\, C & Teichmann\, S.A. (200 6) 3DComplex: a structural classification of protein complexes. PLoS Comp Biol.\, 2\, e155.\n\n \n\n2) Personality and Language in Social Media\n\nI n Facebook\, we studied the relationship between sociometric popularity (n umber of Facebook contacts) and personality traits [1]. We tested to which extent two prevalent viewpoints hold. That is\, sociometrically popular F acebook users (those with many social contacts) are the ones whose persona lity traits either predict many offline (real world) friends or predict pr opensity to maintain superficial relationships. We found that the stronges t predictor for number of friends in the real world (Extraversion) is also the strongest predictor for number of Facebook contacts. We then verified a widely held conjecture that has been put forward by literary intellectu als and scientists alike but has not been tested: people who have many soc ial contacts on Facebook are the ones who are able to adapt themselves to new forms of communication\, present themselves in likeable ways\, and hav e propensity to maintain superficial relationships. We will see that there is no statistical evidence to support such a conjecture. In Twitter\, ins tead\, we tested whether users can be reduced to look-alike nodes (as most of the spreading models would assume) or\, instead\, whether they show in dividual differences that impact their popularity and influence. Again\, o ne aspect that may differentiate users is their character and personality. For 335 users\, we gather personality data\, analyze it\, and find that b oth popular users and influentials are extroverts and emotionally stable ( low in the trait of Neuroticism) [2]. Interestingly\, we also find that po pular users are 'imaginative' (high in Openness)\, while influentials tend to be 'organised' (high in Conscientiousness). We then show a way of accu rately predicting a user's personality simply based on three counts public ly available on profiles: following\, followers\, and listed counts. Knowi ng these three quantities about an active user\, one can predict the user' s five personality traits with a root- mean-squared error below 0.88 on a [1\,5] scale. Also\, since it has been shown that personality is linked to the use of language (which is unobtrusively observable in tweets)\, we ca rry out a study of tweets and show that popular and influential users ling uistically structure their tweets in specific ways [3]. This suggests that the popularity and influence of a Twitter account cannot be simply traced back to the graph properties of the network within which it is embedded\, but also depends on the personality and emotions of the human being behin d it. After introducing these studies\, we will discuss theoretical implic ations of our findings as well as practical implications for viral marketi ng and social media security.\n\n[1] The Personality of Popular Facebook U sers. CSCW 2012. http://dl.dropbox.com/u/6314563/papers/personality_fb.pdf \n\n[2] Our Twitter Profiles\, Our Selves: Predicting Personality with Twi tter. SocialCom 2011. http://www.cl.cam.ac.uk/~dq209/publications/quercia1 1twitter.pdf\n\n[3] In the Mood for Being Influential on Twitter. SocialCo m 2011. http://www.cl.cam.ac.uk/~dq209/publications/quercia11mood.pdf LOCATION:Keynes Hall in Kings College END:VEVENT END:VCALENDAR