FaST linear mixed models for genome-wide association studies

We describe factored spectrally transformed linear mixed models (FaST-LMM), an algorithm for genome-wide association studies (GWAS) that scales linearly with cohort size in both run time and memory use. On Wellcome Trust data for 15,000 individuals, FaST-LMM ran an order of magnitude faster than current efficient algorithms. Our algorithm can analyze data for 120,000 individuals in just a few hours, whereas current algorithms fail on data for even 20,000 individuals (http://mscompbio.codeplex.com/).     

Conditional Random Fields for Fast, Large-Scale Genome-Wide Association Studies

Understanding the role of genetic variation in human diseases remains an important problem to be solved in genomics. An important component of such variation consist of variations at single sites in DNA, or single nucleotide polymorphisms (SNPs). Typically, the problem of associating particular SNPs to phenotypes has been confounded by hidden factors such as the presence of population structure, family structure or cryptic relatedness in the sample of individuals being analyzed. Such confounding factors lead to a large number of spurious associations and missed associations. Various statistical methods have been proposed to account for such confounding factors such as linear mixed-effect models (LMMs) or methods that adjust data based on a principal components analysis (PCA), but these methods either suffer from low power or cease to be tractable for larger numbers of individuals in the sample. Here we present a statistical model for conducting genome-wide association studies (GWAS) that accounts for such confounding factors. Our method scales in runtime quadratic in the number of individuals being studied with only a modest loss in statistical power as compared to LMM-based and PCA-based methods when testing on synthetic data that was generated from a generalized LMM. Applying our method to both real and synthetic human genotype/phenotype data, we demonstrate the ability of our model to correct for confounding factors while requiring significantly less runtime relative to LMMs. We have implemented methods for fitting these models, which are available at http://www.microsoft.com/science.     

Early Selection in Gag by Protective HLA Alleles Contributes to Reduced HIV-1 Replication Capacity That May Be Largely Compensated for in Chronic Infection

Mutations that allow escape from CD8 T-cell responses are common in HIV-1 and may attenuate pathogenesis by reducing viral fitness. While this has been demonstrated for individual cases, a systematic investigation of the consequence of HLA class I-mediated selection on HIV-1 in vitro replication capacity (RC) has not been undertaken. We examined this question by generating recombinant viruses expressing plasma HIV-1 RNA-derived Gag-Protease sequences from 66 acute/early and 803 chronic untreated subtype B-infected individuals in an NL4-3 background and measuring their RCs using a green fluorescent protein (GFP) reporter CD4 T-cell assay. In acute/early infection, viruses derived from individuals expressing the protective alleles HLA-B*57, -B*5801, and/or -B*13 displayed significantly lower RCs than did viruses from individuals lacking these alleles (P < 0.05). Furthermore, acute/early RC inversely correlated with the presence of HLA-B-associated Gag polymorphisms (R = −0.27; P = 0.03), suggesting a cumulative effect of primary escape mutations on fitness during the first months of infection. At the chronic stage of infection, no strong correlations were observed between RC and protective HLA-B alleles or with the presence of HLA-B-associated polymorphisms restricted by protective alleles despite increased statistical power to detect these associations. However, RC correlated positively with the presence of known compensatory mutations in chronic viruses from B*57-expressing individuals harboring the Gag T242N mutation (n = 50; R = 0.36; P = 0.01), suggesting that the rescue of fitness defects occurred through mutations at secondary sites. Additional mutations in Gag that may modulate the impact of the T242N mutation on RC were identified. A modest inverse correlation was observed between RC and CD4 cell count in chronic infection (R = −0.17; P < 0.0001), suggesting that Gag-Protease RC could increase over the disease course. Notably, this association was stronger for individuals who expressed B*57, B*58, or B*13 (R = −0.27; P = 0.004). Taken together, these data indicate that certain protective HLA alleles contribute to early defects in HIV-1 fitness through the selection of detrimental mutations in Gag; however, these effects wane as compensatory mutations accumulate in chronic infection. The long-term control of HIV-1 in some persons who express protective alleles suggests that early fitness hits may provide lasting benefits.The host immune response elicited by CD8⁺ cytotoxic T lymphocytes (CTLs) is a major contributor to viral control following human immunodeficiency virus type 1 (HIV-1) infection (6, 39), but antiviral pressure exerted by CTLs is diminished by the selection of escape mutations in targeted regions throughout the viral proteome (7, 18, 29, 35, 41, 45, 57). A comprehensive identification of HLA-associated viral polymorphisms has recently been achieved through population-based analyses of HIV-1 sequences and HLA class I types from different cohorts worldwide (3, 8, 13-15, 34, 43, 50, 56, 63). However, despite improved characterization of the sites and pathways of immune escape, effective ways to incorporate these findings into immunogen design remain an area of debate. A better understanding of the impact of escape mutations on viral fitness may provide novel directions for HIV-1 vaccines that are designed to attenuate pathogenesis.The development of innovative vaccine strategies that can overcome the extreme diversity of HIV is a key priority (4). One proposed approach is to target the most conserved T-cell epitopes, which presumably cannot escape from CTL pressure easily due to structural or functional constraints on the viral protein (55). Complementary approaches include the design of polyvalent and/or mosaic immunogens that incorporate commonly observed viral diversity (4, 38) or the specific targeting of vulnerable regions of the viral proteome that do escape but only at a substantial cost to viral replication capacity (RC) (1, 40). A chief target of such vaccine approaches is the major HIV-1 structural protein Gag, which is known to be highly immunogenic and to elicit CTL responses that correlate with the natural control of infection (22, 36, 66). Indeed, several lines of evidence support a relationship between the selection of CTL escape mutations and reduced HIV-1 fitness. These include the reversion of escape mutations following transmission to an HLA-mismatched recipient who cannot target the epitope (19, 24, 41) as well as reduced plasma viral load (pVL) set point following the transmission of certain escape variants from donors who expressed protective HLA alleles (17, 27). Notably, these in vivo observations have been made most often for variations within Gag that are attributed to CTL responses restricted by the protective alleles HLA-B*57 and -B*5801 (17, 19, 27, 41). Most recently, reduced in vitro RCs of clinical isolates and/or engineered strains encoding single or multiple escape mutations in Gag selected in the context of certain protective HLA alleles, including B*57, B*5801, B*27, and B*13, have been demonstrated (9, 10, 42, 53, 59, 62). Despite these efforts, the goal of a T-cell vaccine that targets highly conserved and attenuation-inducing sites is hampered by a lack of knowledge concerning the contribution of most escape mutations to HIV-1 fitness as well as a poor understanding of the relative influence of HLA on the viral RC at different stages of infection.The mutability of HIV-1 permits the generation of progeny viruses encoding compensatory mutations that restore normal protein function and/or viral fitness. Detailed studies have demonstrated that the in vitro RC of escape variants in human and primate immunodeficiency viruses can be enhanced by the addition of secondary mutations outside the targeted epitope (10, 20, 52, 59, 65). Thus, vaccine strategies aimed at attenuating HIV-1 must also consider, among other factors, the frequency, time course, and extent to which compensation might overcome attenuation mediated by CTL-induced escape. Despite its anticipated utility for HIV-1 vaccine design, systematic studies to examine the consequences of naturally occurring CTL escape and compensatory mutations on viral RC have not been undertaken.We have described previously an in vitro recombinant viral assay to examine the impact of Gag-Protease mutations on HIV-1 RC (47, 49). Gag and protease have been included in each virus to minimize the impact of sequence polymorphisms at Gag cleavage sites, which coevolve with changes in protease (5, 37). Using this approach, we have demonstrated that viruses derived from HIV-1 controllers replicated significantly less well than those derived from noncontrollers and that these differences were detectable at both the acute/early (49) and chronic (47) stages. Escape mutations in Gag associated with the protective HLA-B*57 allele, as well as putative compensatory mutations outside known CTL epitopes, contributed to this difference in RC (47). However, substantial variability was observed for viruses from controllers and noncontrollers, indicating that additional factors were likely to be involved. Benefits of this assay include its relatively high-throughput capacity as well as the fact that clinically derived HIV-1 sequences are used in their entirety. Thus, it is possible to examine a large number of “real-world” Gag-Protease sequences, to define an RC value for each one, and to identify sequences within the population of recombinant strains that are responsible for RC differences.Here, we use this recombinant virus approach to examine the contribution of HLA-associated immune pressure on Gag-Protease RC during acute/early (n = 66) and chronic (n = 803) infections in the context of naturally occurring HIV-1 subtype B isolates from untreated individuals. In a recent report (64), we employed this system to examine the Gag-Protease RC in a similar cohort of chronic HIV-1 subtype C-infected individuals. The results of these studies provide important insights into the roles of immune pressure and fitness constraints on HIV-1 evolution that may contribute to the rational design of an effective vaccine.     

Design, expression, and processing of epitomized hepatitis C virus-encoded CTL epitopes

Hepatitis C virus (HCV) vaccine efficacy may crucially depend on immunogen length and coverage of viral sequence diversity. However, covering a considerable proportion of the circulating viral sequence variants would likely require long immunogens, which for the conserved portions of the viral genome, would contain unnecessarily redundant sequence information. In this study, we present the design and in vitro performance analysis of a novel "epitome" approach that compresses frequent immune targets of the cellular immune response against HCV into a shorter immunogen sequence. Compression of immunological information is achieved by partial overlapping shared sequence motifs between individual epitopes. At the same time, sequence diversity coverage is provided by taking advantage of emerging cross-reactivity patterns among epitope variants so that epitope variants associated with the broadest variant cross-recognition are preferentially included. The processing and presentation analysis of specific epitopes included in such a compressed, in vitro-expressed HCV epitome indicated effective processing of a majority of tested epitopes, although re-presentation of some epitopes may require refined sequence design. Together, the present study establishes the epitome approach as a potential powerful tool for vaccine immunogen design, especially suitable for the induction of cellular immune responses against highly variable pathogens.     

Phylogenetic Dependency Networks: Inferring Patterns of CTL Escape and Codon Covariation in HIV-1 Gag

HIV avoids elimination by cytotoxic T-lymphocytes (CTLs) through the evolution of escape mutations. Although there is mounting evidence that these escape pathways are broadly consistent among individuals with similar human leukocyte antigen (HLA) class I alleles, previous population-based studies have been limited by the inability to simultaneously account for HIV codon covariation, linkage disequilibrium among HLA alleles, and the confounding effects of HIV phylogeny when attempting to identify HLA-associated viral evolution. We have developed a statistical model of evolution, called a phylogenetic dependency network, that accounts for these three sources of confounding and identifies the primary sources of selection pressure acting on each HIV codon. Using synthetic data, we demonstrate the utility of this approach for identifying sites of HLA-mediated selection pressure and codon evolution as well as the deleterious effects of failing to account for all three sources of confounding. We then apply our approach to a large, clinically-derived dataset of Gag p17 and p24 sequences from a multicenter cohort of 1144 HIV-infected individuals from British Columbia, Canada (predominantly HIV-1 clade B) and Durban, South Africa (predominantly HIV-1 clade C). The resulting phylogenetic dependency network is dense, containing 149 associations between HLA alleles and HIV codons and 1386 associations among HIV codons. These associations include the complete reconstruction of several recently defined escape and compensatory mutation pathways and agree with emerging data on patterns of epitope targeting. The phylogenetic dependency network adds to the growing body of literature suggesting that sites of escape, order of escape, and compensatory mutations are largely consistent even across different clades, although we also identify several differences between clades. As recent case studies have demonstrated, understanding both the complexity and the consistency of immune escape has important implications for CTL-based vaccine design. Phylogenetic dependency networks represent a major step toward systematically expanding our understanding of CTL escape to diverse populations and whole viral genes.     

HLA-Associated Immune Escape Pathways in HIV-1 Subtype B Gag,Pol and Nef Proteins

Background

Despite the extensive genetic diversity of HIV-1, viral evolution in response to immune selective pressures follows broadly predictable mutational patterns. Sites and pathways of Human Leukocyte-Antigen (HLA)-associated polymorphisms in HIV-1 have been identified through the analysis of population-level data, but the full extent of immune escape pathways remains incompletely characterized. Here, in the largest analysis of HIV-1 subtype B sequences undertaken to date, we identify HLA-associated polymorphisms in the three HIV-1 proteins most commonly considered in cellular-based vaccine strategies. Results are organized into protein-wide escape maps illustrating the sites and pathways of HLA-driven viral evolution.

Methodology/Principal Findings

HLA-associated polymorphisms were identified in HIV-1 Gag, Pol and Nef in a multicenter cohort of >1500 chronically subtype-B infected, treatment-naïve individuals from established cohorts in Canada, the USA and Western Australia. At q≤0.05, 282 codons commonly mutating under HLA-associated immune pressures were identified in these three proteins. The greatest density of associations was observed in Nef (where close to 40% of codons exhibited a significant HLA association), followed by Gag then Pol (where ∼15–20% of codons exhibited HLA associations), confirming the extensive impact of immune selection on HIV evolution and diversity. Analysis of HIV codon covariation patterns identified over 2000 codon-codon interactions at q≤0.05, illustrating the dense and complex networks of linked escape and secondary/compensatory mutations.

Conclusions/Significance

The immune escape maps and associated data are intended to serve as a user-friendly guide to the locations of common escape mutations and covarying codons in HIV-1 subtype B, and as a resource facilitating the systematic identification and classification of immune escape mutations. These resources should facilitate research in HIV epitope discovery and host-pathogen co-evolution, and are relevant to the continued search for an effective CTL-based AIDS vaccine.     

Platelet-activating factor and metastasis: calcium-independent phospholipase A2β deficiency protects against breast cancer metastasis to the lung

We determined the contribution of calcium-independent phospholipase A(2)β (iPLA(2)β) to lung metastasis development following breast cancer injection into wild-type (WT) and iPLA(2)β-knockout (iPLA(2)β-KO) mice. WT and iPLA(2)β-KO mice were injected in the mammary pad with 200,000 E0771 breast cancer cells. There was no difference in primary tumor size between WT and iPLA(2)β-KO mice at 27 days postinjection. However, we observed an 11-fold greater number of breast cancer cells in the lungs of WT mice compared with iPLA(2)β-KO animals (P < 0.05). Isolated WT lung endothelial cells demonstrated a significant increase in platelet-activating factor (PAF) production when stimulated with thrombin [1 IU/ml, 10 min, 4,330 ± 555 vs. 15,227 ± 1,043 disintegrations per minute (dpm), P < 0.01] or TNF-α (10 ng/ml, 2 h, 16,532 ± 538 dpm, P < 0.01). Adherence of E0771 cells to WT endothelial cells was increased by thrombin (4.8 ± 0.3% vs. 70.9 ± 6.3, P < 0.01) or TNF-α (60.5 ± 4.3, P < 0.01). These responses were blocked by pretreatment with the iPLA(2)β-selective inhibitor (S)-bromoenol lactone and absent in lung endothelial cells from iPLA(2)β-KO mice. These data indicate that endothelial cell iPLA(2)β is responsible for PAF production and adherence of E0771 cells and may play a role in cancer cell migration to distal locations.