The population of Thailand is nearly 64 million (Ministry of Public Health Thailand, 2010). The most populated region is the North-Eastern region, which has a population of approximately 33.8% of the total Thai population. Khon Kaen University (KKU), one of the national research universities of Thailand, is located in Khon Kaen province, which is a major center in the Northeast. Although a twin registry has not been established in KKU, a few studies of Thai twins have been performed with a forensic science purpose. Consequently, a database of palm prints and fingerprints of those twin subjects is available. Twins are of special interest for genetic studies due to their genetic similarity. Fingerprint pattern (FP) inheritance is under the control of at least seven different genes (Slatis et al., Reference Slatis, Katznelson and Bonne-Tamir1976) coupled with an embryonic environmental factor (Penrose & Ohara, Reference Penrose and Ohara1973). The Thai Twin Registry (TTR) has been established in KKU. Starting with fingerprint research using an existing Thai twin dermatoglyphic database, interesting questions have been raised such as: (1) Are monozygotic (MZ) twins more similar than dizygotic (DZ) twins in dermatoglyphic variables? (2) To what extent do genetic and environmental factors influence dermatoglyphic variables in the Thai population? The present study summarizes how to obtain dermatoglyphic data and the results of dermatoglyphic analyses, and some of those research questions are explored as well.
Aims of the TTR
The ultimate goal of the TTR is to establish a database of Thai twin pairs in which socio-economic information and biological specimens are included. As an initial step, we are currently collecting dermatoglyphic information of palm prints and fingerprints to analyze differences of FP similarities within MZ and DZ twin pairs. Moreover, a selected cutoff point of the FP similarity with same-sex twins is investigated to determine the MZ or the DZ by using the Receiver Operator Characteristic (ROC) and the area under the curve (AUC) analysis (Murphy et al. Reference Murphy, Berwick, Weinstein, Borus, Budman and Klerman1987).
The Database of Possible Twins
It appears that no research has been conducted using twin registries in Thailand. However, the National Science Center for Education, Ministry of Education (http://www.sciplanet.org) held a ‘Twin Meeting Day’ in 2007. A database of that twins’ meeting is not publicly available. There are three sources of information on possible twins: (1) hospitals, (2) the civilian Registrar's Office of each province, and (3) the schools in municipalities and rural areas. Hospitals at the provincial level are the best source for twin birth registrations. While the municipality and the district Civilian Registrar Offices possess birth date and home addresses of twins, schools are one of the best sources of school-aged twins.
Location and Recruitment of the Twin Subjects
Recruitment of twins has been performed using three methods. First, official letters from KKU were distributed to the headmasters of primary and junior high schools for cooperation in developing a twin registry. The second method is called a snowball technique: a friendly talk with schoolteachers or twin subjects has led to recruitment of other eligible twins they knew. The third method, publicity via television news, identified schools in which several twin pairs studied. In total, one province (Suphanburi) in the central region and seven provinces (Khon Kaen, Nakhon-ratchasima, Chaiyaphum, Surin, Sakon Nakhon, Kalasin, and Mahasarakham) located in the Northeast were selected for recruitment of the TTR.
Ethical Issues
The research protocol titled ‘Dermatoglyphic study of Thai twins’ was reviewed and approved by Khon Kaen University Ethics Committee for Human Research (HE532334; IRB00001189; FWA00003418).
Data Collection Procedure
Twin subjects recruited from primary and junior high schools and villages were asked to participate in palm and fingerprinting. In total, 212 twin pairs (155 male, 269 female), aged between 6 to 66 years were informed and gave consent for printing their 10 fingertips and palm prints using a transparent adhesive tape technique (O'Leary et al., Reference O'Leary, Slaney, Bryant and Fraser1986).
Zygosity Determination
The twin subjects were classified into two types, MZ and DZ twins, using two criteria: the biological facts (ABO blood type and sex), and the zygosity questionnaire (Song et al., Reference Song, Lee, Lee, Lee, Lee and Hong2010). The zygosity questionnaire for physical characteristics was modified to fit the Thai culture. Each twin pair was checked whether they were the same sex and same ABO blood type, as well as two items of ‘yes’ responses from the four physical characteristics questions, or else zygosity information determined as MZ twins by a medical doctor, resulting in classification of MZ twins. For the DZ twins determination, same sex but different ABO blood type or differences in sex and ABO blood type, as well as ‘no’ responses to the four physical characteristics questions or zygosity determined as fraternal twins by a medical doctor, were checked.
Database Available
An availability of dermatoglyphic (palm and fingerprints) database of 212 twin pairs was made possible, as it had been kept both in hard copies and digital files by the authors at KKU, Thailand.
Summary of Recent Research
The fingerprint pattern (FP) types were identified by the first author and then classified into four types, that is, arch (A), radial loop (RL), ulnar loop (UL), and whorl (W). The FP similarity for 10 homologous fingers was assessed by matching the same FP type on the homologous fingers between the DZ and the MZ twins. A value of 1 means there was only one homologous finger possessing the same FP type. Thus, the maximal FP similarity was 10, while zero was the minimal value of FP similarity.
The 81 pairs of DZ twin (162 persons) and 131 MZ twin pairs (262 persons) were classified based on the mentioned criteria. The FP frequency and the FP similarity scores were analyzed by DZ and MZ twins to see the magnitude of significant differences between them.
FP Frequency
Frequencies of four FP types by fingers were analyzed comparing DZ and MZ. This revealed that only the right little finger had FP significant difference (LR χ2 (3) = 10.14, p = .017). The A and RL were found as 1.2% and 3.1% for the DZ twins while 0.8% and none for the MZ twins; W was found 24.1% for the DZ twins compared to 26.7% for the MZ twins.
FP Similarity
The MZ twin pairs had an average score of FP similarity of 10 homologous fingers significantly greater than that of the DZ twin, 7.82 (SD = 1.56) vs. 6.43 (2.01).
ROC and AUC Analysis for Twin Differentiation
In medical research, the AUC is commonly used as a statistic calculated on the observed disease scale and is a measure of an efficiency of phenotype prediction using a test classifier (Wray et al., Reference Wray, Yang, Goddard and Visscher2010). For TTR, the ROC was obtained by plotting sensitivities against one minus specificities using the cutoff point of FP similarity of 6 to differentiate the MZ twins from the DZ twins. The AUC was constructed by an SPSS for Windows (version 17). It yielded an AUC of 0.949 (95% CI = 0.901–0.906) indicating a high efficiency for discriminating the MZ twins from the DZ twins (see Figure 1).
FIGURE 1 The ROC curve for the FP similarity.
Conclusion
This is considered to be a valuable study in the collection and analysis of palm and fingerprint analyses of a significant number of twins, both MZ and DZ twins of the Thai people.
As indicated, this is part of ongoing research at the TTR using palm and fingerprint analyses, and is part of the process of developing a Thai twin database at Khon Kaen University, a major government university in northeastern Thailand. The researchers are considered to be a significant authority on this topic in Thailand.
Acknowledgments
We are grateful to twin participants for their cooperation in palm and fingerprints research. We thank Professor Brian Sheehan for his help in English language editing and constructive comments. We wish to thank Drs Yoon-Mi Hur and Jeff Craig for their helpful comments about the earlier version of this manuscript. This research was supported by Khon Kaen University under the Incubation Researcher Project.
