De-anonymizing Genomic Databases Using Phenotypic Traits

Abstract

People increasingly have their genomes sequenced and some of them share their genomic data online. They do so for various purposes, including to find relatives and to help advance genomic research. An individual's genome carries very sensitive, private information such as its owner's susceptibility to diseases, which could be used for discrimination. Therefore, genomic databases are often anonymized. However, an individual's genotype is also linked to visible phenotypic traits, such as eye or hair color, which can be used to re-identify users in anonymized public genomic databases, thus raising severe privacy issues. For instance, an adversary can identify a target's genome using known her phenotypic traits and subsequently infer her susceptibility to Alzheimer's disease. In this paper, we quantify, based on various phenotypic traits, the extent of this threat in several scenarios by implementing de-anonymization attacks on a genomic database of OpenSNP users sequenced by 23andMe. Our experimental results show that the proportion of correct matches reaches 23% with a supervised approach in a database of 50 participants. Our approach outperforms the baseline by a factor of four, in terms of the proportion of correct matches, in most scenarios. We also evaluate the adversary's ability to predict individuals' predisposition to Alzheimer's disease, and we observe that the inference error can be halved compared to the baseline. We also analyze the effect of the number of known phenotypic traits on the success rate of the attack. As progress is made in genomic research, especially for genotype-phenotype associations, the threat presented in this paper will become more serious.

Identification of individuals by trait prediction using whole-genome sequencing data

“…However, phenotype prediction may allow the identification of individuals through genomics-an issue that implicates the privacy of genomic data. Today, where online services with personal images coexist with large genetic databases, such as 23andMe, associating genomic data to physical traits (e.g., eye and skin color) obtains particular relevance (6). In fact, genome data may be linked to metadata through online social networks and services, thus complicating the protection of genome privacy (7).…”

“…Although individually, some of these phenotypes have been evaluated (1,15), we propose an algorithm that integrates each predictive model to match a deidentified WGS sample to phenotypic and demographic information at higher accuracy. When the source of the phenotypic data is of known identity, this procedure may reidentify a genomic sample, raising implications for genomic privacy (6)(7)(8)(9)16).…”

“…In contrast to sN , where a genome is paired to the most similar phenotypic profile, for mN , each genome was paired to one and only one phenotypic set in a globally optimal manner. That is, we treated mN as a bipartite graph matching problem and maximized the expected number of correct pairs (6,43). Table 2 shows sN and mN accuracy across feature sets and pool sizes averaged over all possible lineups per CV fold.…”

Linkage disequilibrium connects genetic records of relatives typed with disjoint genomic marker sets

“…In addition to magnifying the exposure of the relatives of SNP-profile contributors to forensic identification, it also increases the phenotypic reach of STR profiles. CODIS genotypes of one individual could potentially be associated with genomic SNP genotypes of a relative, which could, in turn, reveal phenotypes of that relative [27]. Thus, not only could a CODIS genotype profile reveal phenotypic information about an individual [13], it could also reveal phenotypes of relatives.…”

Inference Attacks against Kin Genomic Privacy

“…Although most share their genomic data on such platforms in an anonymized way, others either directly reveal their real identities or share sufficient information to cause deanonymization. 16,32 By analyzing the genomic data of such websites' users, attackers might be able to infer family bonds; if at least one family member is identifiable or deanonymized, attackers might be able to reconstruct the actual family tree along with their genomic data.…”