Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Gene functional annotation”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 91 records · Page 5Linked to original sources
PubMed2026

Shared genetic architecture between ADHD and intelligence varies across ADHD subtypes.

BACKGROUND: Attention-deficit/hyperactivity disorder (ADHD) is a heterogeneous neurodevelopmental condition frequently accompanied by cognitive difficulties. Although previous genetic studies have demonstrated substantial overlap between ADHD and intelligence, most have treated ADHD as a single phenotype. However, whether this shared genetic architecture differs across ADHD subtypes remains unclear. METHODS: We conducted a genome-wide cross-trait analysis integrating large-scale genome-wide association study (GWAS) datasets of overall ADHD, its subtypes-childhood ADHD, persistent ADHD, and late-diagnosed ADHD-and intelligence (total N > 300,000). Genome-wide genetic correlations, polygenic overlap, local genetic correlations, and variant-level associations between ADHD phenotypes and intelligence were evaluated to characterize their shared genetic architecture. Shared variants were identified through cross-trait enrichment analyses and subsequently mapped to genes for functional annotation and gene-set enrichment. Bidirectional associations were evaluated using two-sample Mendelian randomization with sensitivity analyses. Additional GWAS datasets were used to validate the robustness of shared loci by assessing the consistency of effect directions. RESULTS: All ADHD phenotypes showed significant negative genetic correlations with intelligence (rg ranging from -0.3442 to -0.4205). Despite these modest genome-wide correlations, cross-trait analyses revealed substantial genetic overlap, including polygenic overlap, local genetic correlations, and variant-level associations. We identified 184 loci jointly associated with ADHD traits and intelligence, including 64 novel loci, whereas no shared loci were detected for persistent ADHD under the current analysis. Functional annotation revealed biologically distinct enrichment patterns across subtypes: childhood ADHD loci were linked to early neurodevelopmental processes, while late-diagnosed ADHD loci were enriched in synapse-related and neuronal signaling pathways. Mendelian randomization analyses suggested bidirectional associations, with stronger evidence supporting a directional association from intelligence to ADHD risk. Furthermore, these shared loci showed largely consistent effect directions across additional GWAS datasets, providing support for the robustness of the findings. CONCLUSIONS: The shared genetic architecture between ADHD and intelligence varies across ADHD subtypes, highlighting distinct biological pathways underlying cognitive heterogeneity in ADHD. These findings suggest that the relationship between ADHD liability and general cognitive ability is not uniform across ADHD subtypes and may inform future research on risk stratification and early identification in child and adolescent psychiatry.

Humans↗
PubMed2006

Dense subgraph computation via stochastic search: application to detect transcriptional modules.

MOTIVATION: In a tri-partite biological network of transcription factors, their putative target genes, and the tissues in which the target genes are differentially expressed, a tightly inter-connected (dense) subgraph may reveal knowledge about tissue specific transcription regulation mediated by a specific set of transcription factors-a tissue-specific transcriptional module. This is just one context in which an efficient computation of dense subgraphs in a multi-partite graph is needed. RESULT: Here we report a generic stochastic search based method to compute dense subgraphs in a graph with an arbitrary number of partitions and an arbitrary connectivity among the partitions. We then use the tool to explore tissue-specific transcriptional regulation in the human genome. We validate our findings in Skeletal muscle based on literature. We could accurately deduce biological processes for transcription factors via the tri-partite clusters of transcription factors, genes, and the functional annotation of genes. Additionally, we propose a few previously unknown TF-pathway associations and tissue-specific roles for certain pathways. Finally, our combined analysis of Cardiac, Skeletal, and Smooth muscle data recapitulates the evolutionary relationship among the three tissues.

Algorithms↗
PubMed2004

Gene ontology application to genomic functional annotation, statistical analysis and knowledge mining.

While a massive amount of biomolecular information is increasingly accumulating in different databanks, on the other hand high-throughput technologies are generating a great quantity of data that need to be annotated with the genomic information available, and interpreted. To this aim, the use of specific ontologies can greatly help either in integrating different information stored within heterogeneous databanks, or in identifying and clustering sequence data sharing common characteristics. In the molecular biology domain, the Gene Ontology (GO) is the most developed and widely used ontology. To demonstrate its great utility in the annotation and biological interpretation of gene sets obtained by means of high-throughput experiments, we implemented the web application here described. It enables functional annotations of a given gene set on a genomic scale and across different species. Within our application the annotations provided by the GO vocabulary allow either to easily bind several information from different resources, or to cluster annotated genes according to their biological characteristics. Through the GO structure it is also possible to represent biological concepts with different specificity levels, from very general to very precise concepts. Furthermore, the statistical evaluation of the categorizations provided by the GO annotations enables to highlight the most significant biological characteristics of a gene set, and therefore to mine knowledge from data. Our created tool meets the need to manage a vast quantity of biological data with a simple user interface adapt also for users with limited informatics knowledge, leading them to evaluate the functional significance of experiment's results with graphical views and statistical indexes in a well-known web browser user interface.

Genomics↗
PubMed2006

Identification and analysis of gene families from the duplicated genome of soybean using EST sequences.

BACKGROUND: Large scale gene analysis of most organisms is hampered by incomplete genomic sequences. In many organisms, such as soybean, the best source of sequence information is the existence of expressed sequence tag (EST) libraries. Soybean has a large (1115 Mbp) genome that has yet to be fully sequenced. However it does have the 6th largest EST collection comprised of ESTs from a variety of soybean genotypes. Many EST libraries were constructed from RNA extracted from various genetic backgrounds, thus gene identification from these sources is complicated by the existence of both gene and allele sequence differences. We used the ESTminer suite of programs to identify potential soybean gene transcripts from a single genetic background allowing us to observe functional classifications between gene families as well as structural differences between genes and gene paralogs within families. The identification of potential gene sequences (pHaps) from soybean allows us to begin to get a picture of the genomic history of the organism as well as begin to observe the evolutionary fates of gene copies in this highly duplicated genome. RESULTS: We identified approximately 45,000 potential gene sequences (pHaps) from EST sequences of Williams/Williams82, an inbred genotype of soybean (Glycine max L. Merr.) using a redundancy criterion to identify reproducible sequence differences between related genes within gene families. Analysis of these sequences revealed single base substitutions and single base indels are the most frequently observed form of sequence variation between genes within families in the dataset. Genomic sequencing of selected loci indicate that intron-like intervening sequences are numerous and are approximately 220 bp in length. Functional annotation of gene sequences indicate functional classifications are not randomly distributed among gene families containing few or many genes. CONCLUSION: The predominance of single nucleotide insertion/deletions and substitution events between genes within families (individual genes and gene paralogs) is consistent with a model of gene amplification followed by single base random mutational events expected under the classical model of duplicated gene evolution. Molecular functions of small and large gene families appear to be non-randomly distributed possibly indicating a difference in retention of duplicates or local expansion.

Evolution, Molecular↗
PubMed2004

Automatic annotation of protein motif function with Gene Ontology terms.

BACKGROUND: Conserved protein sequence motifs are short stretches of amino acid sequence patterns that potentially encode the function of proteins. Several sequence pattern searching algorithms and programs exist foridentifying candidate protein motifs at the whole genome level. However, a much needed and important task is to determine the functions of the newly identified protein motifs. The Gene Ontology (GO) project is an endeavor to annotate the function of genes or protein sequences with terms from a dynamic, controlled vocabulary and these annotations serve well as a knowledge base. RESULTS: This paper presents methods to mine the GO knowledge base and use the association between the GO terms assigned to a sequence and the motifs matched by the same sequence as evidence for predicting the functions of novel protein motifs automatically. The task of assigning GO terms to protein motifs is viewed as both a binary classification and information retrieval problem, where PROSITE motifs are used as samples for mode training and functional prediction. The mutual information of a motif and aGO term association is found to be a very useful feature. We take advantage of the known motifs to train a logistic regression classifier, which allows us to combine mutual information with other frequency-based features and obtain a probability of correct association. The trained logistic regression model has intuitively meaningful and logically plausible parameter values, and performs very well empirically according to our evaluation criteria. CONCLUSIONS: In this research, different methods for automatic annotation of protein motifs have been investigated. Empirical result demonstrated that the methods have a great potential for detecting and augmenting information about the functions of newly discovered candidate protein motifs.

Amino Acid Motifs↗
PubMed2002

Analysis of bovine mammary gland EST and functional annotation of the Bos taurus gene index.

Functional genomic studies of the mammary gland require an appropriate collection of cDNA sequences to assess gene expression patterns from the different developmental and operational states of underlying cell types. To better capture the range of gene expression, a normalized cDNA library was constructed from pooled bovine mammary tissues, and 23,202 expressed sequence tags (EST) were produced and deposited into GenBank. Assembly of these EST with sequences in the Bos taurus Gene Index (BtGI) helped to form 5751 of the current 23,883 tentative consensus (TC) sequences. The majority (87%) of these 5751 assemblies contained only one to three mammary-derived EST. In contrast, 18% of the mammary EST assembled with TC sequences corresponding to 12 genes. These results suggest library normalization was only partially effective, because the reduction in EST for genes abundantly transcribed during lactation could be attributed to pooling. For better assessment of novel content in the mammary library and to add to existing annotation of all bovine sequence elements, gene ontology assignments, and comparative sequence analyses against human genome sequence, human and rodent gene indices, and an index of orthologous alignments of genes across eukaryotes (TOGA) were performed, and results were added to existing BtGI annotation. Over 35,000 of the bovine elements significantly matched human genome sequence, and the positions of some alignments (3%) were unique relative to those using human expressed sequences. Because 3445 TC sequences had no significant match with any data set, mammary-derived cDNA clones representing 23 of these elements were analyzed further for expression and novelty. Only one clone met criteria suggesting the corresponding gene was a divergent ortholog or expressed sequence unique to cattle. These results demonstrate that bovine sequence expression data serve as a resource for characterizing mammalian transcriptomes and identifying those genes potentially unique to ruminants.

Animals↗
PubMed1998

Sources of systematic error in functional annotation of genomes: domain rearrangement, non-orthologous gene displacement and operon disruption.

Functional annotation of proteins encoded in newly sequenced genomes can be expected to meet two conflicting objectives: (i) provide as much information as possible, and (ii) avoid erroneous functional assignments and over-predictions. The continuing exponential growth of the number of sequenced genomes makes the quality of sequence annotation a critical factor in the efforts to utilize this new information. When dubious functional assignments are used as a basis for subsequent predictions, they tend to proliferate, leading to "database explosion". It is therefore important to identify the common factors that hamper functional annotation. As a first step towards that goal, we have compared the annotations of the Mycoplasma genitalium and Methanococcus jannaschii genomes produced in several independent studies. The most common causes of questionable predictions appear to be: i) non-critical use of annotations from existing database entries; ii) taking into account only the annotation of the best database hit; iii) insufficient masking of low complexity regions (e.g. non-globular domains) in protein sequences, resulting in spurious database hits obscuring relevant ones; iv) ignoring multi-domain organization of the query proteins and/or the database hits; v) non-critical functional inferences on the basis of the functions of neighboring genes in an operon; vi) non-orthologous gene displacement, i.e. involvement of structurally unrelated proteins in the same function. These observations suggest that case by case validation of functional annotation by expert biologists remains crucial for productive genome analysis.

Databases, Factual↗
PubMed2006

Modelling in molecular biology: describing transcription regulatory networks at different scales.

Approaches to describe gene regulation networks can be categorized by increasing detail, as network parts lists, network topology models, network control logic models or dynamic models. We discuss the current state of the art for each of these approaches. We study the relationship between different topology models, and give examples how they can be used to infer functional annotations for genes of unknown function. We introduce a new simple way of describing dynamic models called finite state linear model (FSLM). We discuss the gap between the parts list and topology models on one hand, and network logic and dynamic models, on the other hand. The first two classes of models have reached a genome-wide scale, while for the other model classes high-throughput technologies are yet to make a major impact.

Computational Biology↗
PubMed2026

Whole-genome sequencing and analysis of the endophytic fungus Alternaria alternata Y-2 from Leymus chinensis.

To explore the genetic basis and functional potential of beneficial symbiosis between the endophytic fungus Alternaria alternata Y-2 and its host Leymus chinensis, we performed Illumina-based draft whole-genome sequencing and systematic bioinformatic analysis. Although this assembly does not reach telomere-to-telomere completeness, it provides high-quality gene-level information for gene prediction, functional annotation, carbohydrate-active enzyme (CAZyme) identification, and secondary metabolite biosynthetic gene cluster analysis. The final genome size of A. alternata Y-2 was 34,383,676 bp with a GC content of 51.0%, containing 12,724 predicted protein-coding genes, 90 tRNAs, and 12 rRNAs. BUSCO assessment showed 98.9% completeness, supporting the high quality of this draft genome. A total of 12,627 genes were successfully annotated in the NCBI NR database, and 17,183 genes were functionally categorized using GO terms. In total, 448 CAZyme genes and 21 secondary metabolite biosynthetic gene clusters were identified, which are potentially involved in lignocellulose degradation, cellular redox homeostasis and biosynthesis of bioactive metabolites. Based on ITS sequence alignment, NR annotation, and phylogenetic analysis of single-copy orthologous genes, the strain was confidently identified as A. alternata. This study firstly reports the draft genome of an endophytic A. alternata strain derived from L. chinensis and provides valuable genetic resources for exploring the endophytic lifestyle, stress tolerance, and bioactive metabolite potential of this fungus.

Alternaria↗
PubMed2025

Quantitative proteomic analysis of the brain reveals the potential antidepressant mechanism of Jiawei Danzhi Xiaoyao San in a chronic unpredictable mild stress mouse model of depression.

OBJECTIVE: To reveal the antidepressant mechanisms of Jiawei DanZhiXiaoYaoSan (,JD) in chronic unpredictable mild stress (CUMS)-induced depression in mice. METHODS: Using the CUMS mouse model of depression, the antidepressant effects of JD were assessed using the sucrose preference test (SPT), forced swimming test (FST), and tail suspension test (TST). Tandem mass tag (TMT)-based quantitative proteomic analysis of the brain was performed following JD treatment. Hierarchical clustering, Gene Ontology function annotation, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment, and protein-protein interactions (PPIs) were used to analyze differentially expressed proteins (DEPs), which were further validated using quantitative real-time polymerase chain reaction (qRT-PCR) and Western blotting. RESULTS: Behavioral tests confirmed the anti-depressant effects of JD, and bioinformatics analysis revealed 59 DEPs, including 33 up-regulated and 26 down-regulated proteins, between the CUMS and JD-M groups. KEGG and PPI analyses revealed that neuro-filament proteins and the Ras signaling pathway may be key targets of JD in the treatment of depression. qRT-PCR and Western blotting results demonstrated that CUMS reduced the protein expression of neurofilament light (NEFL) and medium (NEFM) and inhibited the phosphorylation of extracellular regulated kinase 1/2 (ERK1/2), whereas JD promoted the phosphorylation of ERK1/2 and up-regulated the protein expression of NEFL and NEFM. CONCLUSIONS: The antidepressant mechanism of JD may be related to the up-regulation of p-ERK1/2 and neurofilament proteins.

Animals↗
PubMed2005

Microarray annotation and biological information on function.

OBJECTIVES: Many methods for statistical analysis of gene expression studies by DNA microarrays produce lists of genes as output. To understand gene lists in terms of traditional biology, e.g. which pathways may be affected, it is necessary to get appropriate annotations for the probes on an array. METHODS: Problems arise with the different sources that have been used by manufacturers to design microarray probes, and their association to biological entities like genes, transcripts and proteins. Function annotation is of crucial importance, and systems like Gene Ontology can be used for this purpose. It arranges annotation terms in a hierarchical manner and thus makes annotations in a gene list amenable to automated analysis. RESULTS: Several methods for analyses of gene function are described. The hierarchical nature of systems like Gene Ontology particularly suggests using methods from graph theory. CONCLUSIONS: The main problem in annotating microarray probes and inferring affected functional modules is the incompleteness and degree of error in current biological databases. Initial approaches to make use of functional annotation exist, but have to be extended, in particular with respect to estimating the statistical significance of results.

Computational Biology↗
PubMed2025

Fishing for a reelGene: evaluating gene models with evolution and machine learning.

Assembled genomes and their associated annotations have transformed our study of gene function. However, each new annotated assembly generates new gene models. Inconsistencies between annotations likely arise from biological and technical causes, including pseudogene misclassification, transposon activity, and intron retention from sequencing of unspliced transcripts. To evaluate gene model predictions, we developed reelGene, a pipeline of machine learning models focused on (1) transcription boundaries, (2) mRNA integrity, and (3) protein structure. The first two models leverage sequence characteristics and evolutionary conservation across related taxa to learn the grammar of conserved transcription boundaries and mRNA sequences, while the third uses the conserved evolutionary grammar of protein sequences to predict whether a gene can produce a protein. Evaluating 1.8 million transcript models in Zea mays ssp. mays (maize), reelGene classified 28% as incorrectly annotated or non-functional. We find that reelGene classifies 92.2% of genes in the maize proteome and 99.2% of genes within the maize classical gene list as functional. reelGene also provides a way to further investigate genome biology- for instance, reelGene indicates that 10.3% of dispensable genes in B73 are functional, and within retained duplicate genes, reelGene identifies a 30% bias toward the retention of the M1 subgenome when one copy is functional and the other is non-functional. As an annotation-evaluating tool, reelGene is directly applicable to species of the Andropogoneae tribe, including other important crops like sorghum and miscanthus. As a community resource, reelGene has been integrated onto MaizeGDB both as a browser track and as an individual Shiny App, allowing researchers to evaluate gene model accuracy and further investigate genome biology.

Machine Learning↗
PubMed2005

Phylogenetic detection of conserved gene clusters in microbial genomes.

BACKGROUND: Microbial genomes contain an abundance of genes with conserved proximity forming clusters on the chromosome. However, the conservation can be a result of many factors such as vertical inheritance, or functional selection. Thus, identification of conserved gene clusters that are under functional selection provides an effective channel for gene annotation, microarray screening, and pathway reconstruction. The problem of devising a robust method to identify these conserved gene clusters and to evaluate the significance of the conservation in multiple genomes has a number of implications for comparative, evolutionary and functional genomics as well as synthetic biology. RESULTS: In this paper we describe a new method for detecting conserved gene clusters that incorporates the information captured by a genome phylogenetic tree. We show that our method can overcome the common problem of overestimation of significance due to the bias in the genome database and thereby achieve better accuracy when detecting functionally connected gene clusters. Our results can be accessed at database GeneChords http://genomics10.bu.edu/GeneChords. CONCLUSION: The methodology described in this paper gives a scalable framework for discovering conserved gene clusters in microbial genomes. It serves as a platform for many other functional genomic analyses in microorganisms, such as operon prediction, regulatory site prediction, functional annotation of genes, evolutionary origin and development of gene clusters.

Algorithms↗
PubMed2026

Characterization of a draft chromosome-scale genome assembly for the mutton snapper, Lutjanus analis.

BACKGROUND: The mutton snapper (Lutjanus analis) is a reef fish commonly found in tropical waters of the Western Atlantic Ocean. Genomic studies of this species are needed to support conservation efforts and breeding programs. OBJECTIVE: Here, we report the development of a chromosome-scale reference assembly for the mutton snapper and conduct an initial comparative genomic analysis with other lutjanids. METHODS: The genome of one mutton snapper specimen was sequenced using PAC-Bio HiFi long reads and Illumina short reads. Contigs and scaffolds were assembled in the Flye pipeline and anchored using Hi-C proximity guided assembly. Gene prediction and functional annotations were obtained in AUGUSTUS and eggNOG-mapper, respectively. The mutton snapper genome was compared to those of other lutjanids to infer gene family evolution and chromosome synteny conservation. RESULTS: Assembly and polishing yielded 946 contigs and 926 scaffolds (N50 of 3.16 Mb, complete BUSCO score 98.1%) that were anchored using Hi-C scaffolding in 24 draft chromosomes. The anchored assembly featured a N50 of 42.47 Mb and contained 97.6% of the unanchored assembly length. The 24 mutton snapper chromosomes showed a one-to-one syntenic relationship with their counterparts in medaka, and other Lutjanids. AUGUSTUS predicted 29,023 genes, 24,335 of which (83.85%) could be functionally annotated. Gene family evolution analysis revealed 1,014 significantly expanded or contracted hierarchical ortholog groups in mutton snapper. Expansions and contractions were linked to several biological functions including growth, oocyte maturation, and response to exogenous stressors. CONCLUSION: The draft genome will be a valuable tool for forthcoming applied genomic studies of mutton snapper.

Animals↗
PubMed2006

Gene function classification using Bayesian models with hierarchy-based priors.

BACKGROUND: We investigate whether annotation of gene function can be improved using a classification scheme that is aware that functional classes are organized in a hierarchy. The classifiers look at phylogenic descriptors, sequence based attributes, and predicted secondary structure. We discuss three Bayesian models and compare their performance in terms of predictive accuracy. These models are the ordinary multinomial logit (MNL) model, a hierarchical model based on a set of nested MNL models, and an MNL model with a prior that introduces correlations between the parameters for classes that are nearby in the hierarchy. We also provide a new scheme for combining different sources of information. We use these models to predict the functional class of Open Reading Frames (ORFs) from the E. coli genome. RESULTS: The results from all three models show substantial improvement over previous methods, which were based on the C5 decision tree algorithm. The MNL model using a prior based on the hierarchy outperforms both the non-hierarchical MNL model and the nested MNL model. In contrast to previous attempts at combining the three sources of information in this dataset, our new approach to combining data sources produces a higher accuracy rate than applying our models to each data source alone. CONCLUSION: Together, these results show that gene function can be predicted with higher accuracy than previously achieved, using Bayesian models that incorporate suitable prior information.

Bayes Theorem↗
PubMed2005

Identification of candidate genes for drought stress tolerance in rice by the integration of a genetic (QTL) map with the rice genome physical map.

Genetic improvement for drought stress tolerance in rice involves the quantitative nature of the trait, which reflects the additive effects of several genetic loci throughout the genome. Yield components and related traits under stressed and well-water conditions were assayed in mapping populations derived from crosses of AzucenaxIR64 and AzucenaxBala. To find the candidate rice genes underlying Quantitative Trait Loci (QTL) in these populations, we conducted in silico analysis of a candidate region flanked by the genetic markers RM212 and RM319 on chromosome 1, proximal to the semi-dwarf (sd1) locus. A total of 175 annotated genes were identified from this region. These included 48 genes annotated by functional homology to known genes, 23 pseudogenes, 24 ab initio predicted genes supported by an alignment match to an EST (Expressed sequence tag) of unknown function, and 80 hypothetical genes predicted solely by ab initio means. Among these, 16 candidate genes could potentially be involved in drought stress response.

Disasters↗
PubMed2026

A Network Pharmacology and Molecular Docking Study of TongBi Formula for Osteoarthritis.

This study applied network pharmacology combined with molecular docking to predict the potential therapeutic targets and molecular mechanisms of TongBi Formula (TBF) in osteoarthritis (OA). Active components and corresponding targets of TBF were retrieved from the traditional Chinese medicine Systems Pharmacology Database and Analysis Platform, while OA-related targets were collected from Online Mendelian Inheritance in Man, GeneCards, DrugBank, and Therapeutic Target Database. A network visualization and analysis software was used to construct compound-target and protein-protein interaction (PPI) networks. Gene Ontology functional annotation and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analyses were performed using the Database for Annotation, Visualization and Integrated Discovery platform. Molecular docking analysis was conducted using a molecular docking software to evaluate the predicted binding affinity between key active compounds and core target proteins. A total of 47 overlapping targets between TBF and OA were identified. PPI network analysis highlighted JUN, RELA, IL6, MAPK1, and IL10 as potential hub targets. Enrichment analysis suggested that TBF may regulate inflammation, lipid metabolism, and multiple intracellular signaling pathways associated with OA progression. Molecular docking results demonstrated favorable predicted binding affinities between core active compounds and key OA-related protein targets. These findings provide a computational framework for understanding the potential mechanisms of TBF against OA and support further experimental validation.

Molecular Docking Simulation↗
PubMed2006

Systematic gene function prediction from gene expression data by using a fuzzy nearest-cluster method.

BACKGROUND: Quantitative simultaneous monitoring of the expression levels of thousands of genes under various experimental conditions is now possible using microarray experiments. However, there are still gaps toward whole-genome functional annotation of genes using the gene expression data. RESULTS: In this paper, we propose a novel technique called Fuzzy Nearest Clusters for genome-wide functional annotation of unclassified genes. The technique consists of two steps: an initial hierarchical clustering step to detect homogeneous co-expressed gene subgroups or clusters in each possibly heterogeneous functional class; followed by a classification step to predict the functional roles of the unclassified genes based on their corresponding similarities to the detected functional clusters. CONCLUSION: Our experimental results with yeast gene expression data showed that the proposed method can accurately predict the genes' functions, even those with multiple functional roles, and the prediction performance is most independent of the underlying heterogeneity of the complex functional classes, as compared to the other conventional gene function prediction approaches.

Algorithms↗