Abstract

   Intelligence quotient (IQ) is the most widely used phenotype to
   characterize human cognitive abilities. Recent advances in studies on
   human intelligence have identified many new susceptibility genes.
   However, the genetic mechanisms involved in IQ score and the
   relationship between IQ score and the risk of mental disorders have won
   little attention. To address the genetic complexity of IQ score, we
   have developed IQdb ([32]http://IQdb.cbi.pku.edu.cn), a publicly
   available database for exploring IQ-associated human genes. In total,
   we collected 158 experimental verified genes from literature as a core
   dataset in IQdb. In addition, 46 genomic regions related to IQ score
   have been curated from literature. Based on the core dataset and 46
   confirmed linked genomic regions, more than 6932 potential IQ-related
   genes are expanded using data of protein–protein interactions. A
   systematic gene ranking approach was applied to all the collected and
   expanded genes to represent the relative importance of all the 7090
   genes in IQdb. Our further systematic pathway analysis reveals that
   IQ-associated genes are significantly enriched in multiple signal
   events, especially related to cognitive systems. Of the 158 genes in
   the core dataset, 81 are involved in various psychotic and mental
   disorders. This comprehensive gene resource illustrates the importance
   of IQdb to our understanding on human intelligence, and highlights the
   utility of IQdb for elucidating the functions of IQ-associated genes
   and the cross-talk mechanisms among cognition-related pathways in some
   mental disorders for community.

   Database URL: [33]http://IQdb.cbi.pku.edu.cn.

Introduction

   Human intelligence refers to a set of cognitive abilities, such as
   thinking, remembering, reading, learning, problem solving and using
   language. The high genetic heterogeneity of intelligence poses an
   enormous challenge for understanding molecular mechanisms for
   cognition. Intelligence quotient (IQ) is the most widely used phenotype
   for characterizing human intelligence in psychometric studies. It is
   not surprising that IQ score is consistently associated with a number
   of mental disorders such as schizophrenia, autism, depression and
   anxiety ([34]1–3). Although genetic epidemiology of the relationship
   between IQ score and the risk of related mental disorders becomes
   increasingly clear with various lines of studies, there are no
   substantial achievements to contribute to understanding the molecular
   mechanisms underlying human intelligence and relevant mental disorders.

   As a quantitative trait, the heritability behind an observed IQ score
   is due to complex genetic interactions between multiple genes of small
   effect sizes ([35]4–6). Genetic association studies have identified
   many candidate genes for human intelligence; however, many candidates
   fail to be replicated between studies and populations ([36]4).
   Additionally, current genetic predisposition information is scattered
   in literature and, to date, there has been no systematic collection and
   analysis. Hence, there is no detailed investigation on the common
   molecular mechanisms between IQ score and the risk of related mental
   disorder. Development of a more comprehensive gene resource is really
   desired to gain a more complete molecular picture for intelligence and
   relevant disorders.

   In this article, we present the IQdb, an IQ-associated gene database
   for ongoing development of genes relevant to intelligence and serving
   as a reference dataset for understanding the mechanisms of human
   intelligence. The resultant gene list, preferably in IQdb with
   additional functional and genetic information, including gene
   association study, family-based linkage study, genome-wide association
   study and other functional studies, would be a valuable resource for
   the community. In addition, our systematic pathway and disease
   enrichment analyses reveal that the IQ-associated genes enriched in
   multiple signal events are involved with many cancers and mental
   disorders. To the best of our knowledge, IQdb is the first example of
   an integrated and comprehensive gene resource that helps to elucidate
   the relationship between IQ score and genetic risk factors in mental
   disorders. Our collection could have profound implications for the
   diagnosis, treatment and prevention of some intelligence-related mental
   disorders.

Data Annotations

Collection of core dataset, experimental verified candidate genes

   As shown in [37]Figure 1, this comprehensive collection of gene and
   genomic information for IQdb was accomplished by curating from
   published articles using the following four steps:
    1. An extensive literature search, particularly concerning
       family-based linkage studies, population association studies,
       genome-wide association studies and other functional analyses, was
       conducted through PubMed (on 10 January 2013) using the following
       search terms: [“intelligence quotient” (Title/Abstract) OR “IQ”
       (Title/Abstract)] AND [“genome-wide association study”
       (Title/Abstract) OR “genome wide association study”
       (Title/Abstract)] OR [“gene” (Title/Abstract) OR “genetic”
       (Title/Abstract)] OR [“association” (Title/Abstract) OR “linkage”
       (Title/Abstract)]).
    2. The retrieved 2307 abstracts were highlighted with query keywords
       and grouped by the function in Entrez system in Related Articles.
    3. The 2307 abstracts were read manually to curate the experimental
       verified candidate genes, single-nucleotide polymorphisms (SNPs)
       and genomic regions relevant to IQ and other related information
       such as experimental methods and studied population.
    4. All the names of experimental verified candidate gene and SNPs were
       manually mapped to 158 Entrez Gene IDs and 139 SNP IDs. For
       accuracy, we excluded all negative reports. Finally, we defined the
       158 genes as a core dataset with high confidence. In addition, 46
       genomic regions were also curated from linkage studies ([38]4). To
       expand the IQ-associated gene list, we overlapped the genes to
       these curated 46 genomic regions based on RefSeq gene annotation
       from UCSC genome browser ([39]7).

Figure 1.

   [40]Figure 1.
   [41]Open in a new tab

   Pipeline for collection, expansion and annotation of IQ-associated
   genes.

Expanding and ranking candidate genes from genomic regions and
protein–protein interactions

   The molecular basis underlying IQ score is still unclear because of its
   high genetic heterogeneity. Classical identification of candidate genes
   in individual studies often focuses on verifying specific
   genes/variants predisposing to IQ. Therefore, systematic evaluation and
   summary of relationship between all candidate genes is rare. In this
   article, we first expanded the IQ-associated genes based on the core
   dataset using linked genomic regions and protein–protein interactions.
   Using a multi-dimensional evidence-based candidate gene prioritization
   approach ([42]8), the relative importance of each expanded gene was
   estimated based on the supported evidence from literature, genomics
   regions and functional roles. For instance, 3898 genes locating in the
   46 curated genomic regions were expanded. And 3063 genes that
   interacted with 158 genes in the core dataset were further introduced
   from the BioGRID ([43]9), HPRD ([44]10) and BIND ([45]11) databases.
   Finally, 7090 genes, including the genes in the core dataset, were
   integrated together as a most comprehensive IQ-associated gene list.

   To calculate the relativities of all 7090 genes, a benchmark dataset
   including 19 IQ-associated genes with positive evidence was compiled
   from a classical review ([46]4) ([47]Supplementary File 1). Then, we
   followed a gene prioritization approach ([48]12) to generate a
   candidate weight matrix pool including d^N = 4^3 weight vectors, where
   N represents the number of evidence, including literature, linkage
   regions and interactions, and d = N + 1 represents possible different
   weights, from 1 to 4 in the weight vectors. A combined score for each
   gene was then calculated by summing up the products of the scores and
   the corresponding weights from the three evidence ([49]8). All the 7090
   candidate genes, including 19 benchmark genes, were sorted by their
   combined scores. We selected the optimal weight matrix [4, 1, 1] that
   gave the 95% benchmark genes the highest rank among the top 5% of all
   candidate genes. Based on the matrix, we evaluated the relevance of the
   7090 introduced genes with IQ score, which was useful for users to get
   potential genes for further screening.

Biological function annotations

   Extensive functional information has been retrieved and integrated for
   better understanding the function of the IQ-associated genes, such as
   cross-links to NCBI Entrez gene ([50]13), OMIM ([51]13), UniProt
   ([52]14), Ensembl ([53]15) and Gene Ontology ([54]16). Comprehensive
   mRNA expression profiling data are also collected from BioGPS ([55]17),
   Allen Brain Atlas ([56]18) and RNA-Seq ([57]19–24). Several popular
   pathway databases are used to get comprehensive pathway-related
   information, including BioCyc ([58]25), KEGG Pathway ([59]26), PID
   Curated ([60]27), PANTHER ([61]28), PID Reactome ([62]29, [63]30),
   rate-limiting enzyme database ([64]31), pathway localization database
   ([65]32) and transporter substrate database ([66]33). Other possible
   association diseases are also integrated from GAD (gene association
   database) ([67]34), KEGG Disease ([68]35), FunDO ([69]36, [70]37),
   NHGRI ([71]38) and OMIM ([72]13). In addition, the original IQ-related
   literature references in the NCBI PubMed database are linked to each