General Method to Unravel Ancient Population Structures through Surnames, Final Validation on Italian Data

Abstract

We analyze the geographic location of 77,451 different Italian surnames (17,579,891 individuals) obtained from the lists of telephone subscribers of the year 1993.By using a specific neural network analysis (Self-Organizing Maps, SOMs), we automatically identify the geographic origin of 49,117 different surnames. To validate the methodology, we compare the results to a study, previously conducted, on the same database, with accurate supervised methods. By comparing the results, we find an overlap of 97%, meaning that the SOMs methodology is highly reliable and well traces back the geographic origin of surnames at the time of their introduction (Late Middle Ages/Renaissance in Italy).SOMs results enables one to distinguish monophyletic surnames from polyphyletic ones, that is surnames having had a single geographic and historic origin from those that started to be in use, with an identical spelling, in different locations (respectively, 76.06% and 21.05% of the total). As we are interested in geographic origins, polyphyletic surnames are excluded from further analyses.By comparing the present location of each monophyletic surname to its inferred geographic origin in late Middle Ages/Renaissance, we measure the extent of the migrations having occurred in Italy since that time. We find that the percentage of individuals presently living in the very area where their surname started to be in use centuries ago is extremely variable (ranging from 22.77% to 77.86% according to the province), thus meaning that self-assessed regional identities seldom correspond to the "autochthony" they imply. For example the upper part of the Thyrennian coast (Northern Latium, Tuscany) has a strong identity but few "autochthonous" inhabitants (28%) having been a passageway from the North to the South of Italy. KeywordsSurnames, Family Names, Y-Chromosome, Italy, Migrations, Population Genetics, Demography Cover Page FootnoteThe lists of the surnames autochthonous of each province can be made available to third parties, for scientific purposes, upon request to the corresponding author. By using a specific neural network analysis (Self-Organizing Maps, SOMs), we automatically identify the geographic origin of 49,117 different surnames. To validate the methodology, we compare the results to a study, previously conducted, on the same database, with accurate supervised methods. By comparing the results, we find an overlap of 97%, meaning that the SOMs methodology is highly reliable and well traces back the geographic origin of surnames at the time of their introduction (Late Middle Ages/Renaissance in Italy).SOMs results enables one to distinguish monophyletic surnames from polyphyletic ones, that is surnames having had a single geographic and historic origin from those that started to be in use, with an identical spelling, in different locations (respectively, 76.06% and 21.05% of the total). As we are interested in geographic origins, polyphyletic surnames are excluded from further analyses.By comparing the present location of eac...

Traces of medieval migrations in a socially stratified population from Northern Italy. Evidence from uniparental markers and deep-rooted pedigrees

“…To best reconstruct the paternal genetic variability of the Partecipanza, we sampled at least one individual for each of the 38 Partecipanza surnames. As for the SGP sample, all the selected individuals share surnames that were identified as autochthonous for the considered area by Boattini et al (2012). To avoid any surname-related bias, all SGP individuals bear different surnames (with the negligible exception of three surnames that are represented by two individuals each).…”

“…As a second point, surnames and/or pedigrees can be used for selecting samples within populations. Sampling males bearing surnames that are unequivocally associated with a certain place for Y-chromosome studies provides proxies for older populations (King and Jobling, 2009a;Boattini et al, 2012). The same considerations hold for sample selection based on genealogies/pedigrees.…”

The Italian genome reflects the history of Europe and the Mediterranean basin

“…surnames on the basis of previous works by Zei et al 16,17 Each subject included in the study had a well-defined geographical origin: four grandparents (and parents) born in the same administrative region, assessed through interviews at the time of blood collection. The distribution of the study samples across Italy and the sampling provinces is shown in Figure 1 and Supplementary Table S1 with sample sizes.…”

Complex interplay between neutral and adaptive evolution shaped differential genomic background and disease susceptibility along the Italian peninsula

“…For this purpose, 780 individuals were selected to be representative of variation observable at 20 provinces equally distributed in four geographical macro-areas (i.e. Northern Italy, N_ITA; Central Italy, C_ITA; Southern Italy, S_ITA, and Sardinia, SARD) for which previous studies suggested relatively high internal historical/cultural homogeneity325. In addition to ~280,000 genome-wide SNPs, our survey included also ~250,000 exomic markers and ~20,000 variants at coding/regulatory chromosomal intervals with well-established clinical relevance.…”