Home | About | Journals | Submit | Contact Us | Français |

**|**BMC Bioinformatics**|**v.13; 2012**|**PMC3598729

Formats

Article sections

- Abstract
- Background
- Methods
- Results and discussion
- Conclusion
- Competing interests
- Authors’ contributions
- Supplementary Material
- References

Authors

Related links

BMC Bioinformatics. 2012; 13: 327.

Published online 2012 December 8. doi: 10.1186/1471-2105-13-327

PMCID: PMC3598729

Tahir Mehmood: on.bmu@doomhem.rihat; Jonas Warringer: es.ug.bmc@regnirraw.sanoj; Lars Snipen: on.bmu@nepins.sral; Solve Sæbø: on.bmu@obas.evlos

Received 2012 April 24; Accepted 2012 December 5.

Copyright ©2012 Mehmood et al; licensee BioMed Central Ltd.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This article has been cited by other articles in PMC.

Multivariate approaches have been successfully applied to genome wide association studies. Recently, a Partial Least Squares (PLS) based approach was introduced for mapping yeast genotype-phenotype relations, where background information such as gene function classification, gene dispensability, recent or ancient gene copy number variations and the presence of premature stop codons or frameshift mutations in reading frames, were used *post hoc* to explain selected genes. One of the latest advancement in PLS named L-Partial Least Squares (L-PLS), where ‘L’ presents the used data structure, enables the use of background information at the modeling level. Here, a modification of L-PLS with variable importance on projection (VIP) was implemented using a stepwise regularized procedure for gene and background information selection. Results were compared to PLS-based procedures, where no background information was used.

Applying the proposed methodology to yeast *Saccharomyces cerevisiae* data, we found the relationship between genotype-phenotype to have improved understandability. Phenotypic variations were explained by the variations of relatively stable genes and stable background variations. The suggested procedure provides an automatic way for genotype-phenotype mapping. The selected phenotype influencing genes were evolving 29% faster than non-influential genes, and the current results are supported by a recently conducted study. Further power analysis on simulated data verified that the proposed methodology selects relevant variables.

A modification of L-PLS with VIP in a stepwise regularized elimination procedure can improve the understandability and stability of selected genes and background information. The approach is recommended for genome wide association studies where background information is available.

The explosive growth of data describing the natural genetic and phenotypic variation within species, and the corresponding emergence of populations genomic [1] and population phenomics [2] as nascent fields of research demands new or improved methods for exploring genotype-phenotype relationships. In a recent proof of concept study, we introduced multivariate analysis in the form of Soft-Thresholding Partial Least Squares (ST-PLS) [3] for the mapping of genotype and phenotype interactions in the yeast, *Saccharomyces cerevisiae*[4], which has been at the center point for this development. Multivariate approaches have the potential to provide superior statistical power, increased interpretability of results and a deeper functional understanding of the genotype-phenotype landscape as it pays attention to relationships between multiple genotypes and multiple phenotypes, without producing an excessive number of hypotheses to test. Hence, it could provide decisive advantages over classical univariate analysis [5-9]. A caveat is the sensitivity of multivariate approaches to parameter estimation and this remains a serious challenge, partially because variables tend to show extensive collinearity, which can destroy the asymptotic consistency of the PLS estimators for univariate responses [10] and partially because signal-to-noise ratios are often low. A possible solution to the challenge of estimating parameters correctly is to guide these estimations by including background information on variable relationships at the modeling stage [11]. For genotype-phenotype mapping, such information could encompass the location of genes in the genome, the degree of shared regulatory elements, functional relatedness in terms of biochemical activity, molecular process or subcellular localization of gene products or physical interactions between gene-products. Also, for biological interpretation of the results, focus in relational studies is getting shifted from the selection of genes towards the selection of groups of genes listed as background information [12,13], but the selection of groups of genes can be missed if only few of the corresponding genes are significant [14]; hence a powerful structure extraction tool is required.

Indeed, the analytical implications of including background information to optimize parameter estimation in multivariate analysis, in the context of a three block Partial Least Squares based method named L-PLS regression, has recently been considered [15,16]. The mapping of genotype-phenotype relations through L-PLS requires a variable selection step. We recently [17] introduced a backward stepwise elimination procedure in PLS for identifying codon variations discriminating different bacterial taxa, where significant number of variables were eliminated at the cost of a marginal decrease in model accuracy. This approach was found to be superior to other multivariate procedures with respect to the understandability of the model and the consistency of the estimates. Here, we investigate whether inclusion of background information in the parameter estimation step of a multivariate analysis can increase the stability and understandability when applied in the context of mapping genotype-phenotype relationships in large data sets. Applying L-PLS and a stepwise backward elimination procedure, we find the inclusion of background information to enhance both stability and understandability in an automatic way and thus to constitute a promising way forward.

To demonstrate the efficacy of the proposed procedure for variable selection when background information on the variables is available, simulation data from the following known model, also used by [15], is considered. *E*(** y**)=

$$\phantom{\rule{-12.0pt}{0ex}}h=\left[\begin{array}{l}x\\ \stackrel{~}{y}\end{array}\right]~\mathrm{MVN}\left(\left[\begin{array}{l}0\\ 0\end{array}\right],\left[\begin{array}{ll}{\Sigma}_{x}& {\sigma}_{x\stackrel{~}{y}}\\ {\sigma}_{x\stackrel{~}{y}}^{}1\end{array}\right],=,\mathrm{MVN},(,0,,,\Sigma ,)\right)$$

(1)

where ${\sigma}_{x\stackrel{~}{y}}$ is the covariances vector between ** x** and $\stackrel{~}{y}$. By imposing a block diagonal structure on

$${\Sigma}_{x}=\left[\begin{array}{lllll}{\Sigma}_{{k}_{1}}& 0& \xb7& \xb7& 0\\ 0& {\Sigma}_{{k}_{2}}& \xb7& \xb7& 0\\ \xb7& \xb7& \xb7\\ \xb7& \xb7& \xb7\\ 0& 0& \xb7& \xb7& {\Sigma}_{{k}_{L}}\end{array}\right]$$

(2)

Suppose *k*_{l} is the number of variables in group *l* (*l*=1,…,*L*). A total of *L*=14 groups of variables were simulated with group sizes *k*_{1},…,*k*_{14}=[ 10,10,10,100,10,10,10,10,10,10,10,100,100,100 ], and uniform correlation structures in each ${\Sigma}_{{k}_{l}}$ were assumed internally with correlations equal to *ρ*_{1},…,*ρ*_{14}=[0*.*5,0*.*5,0*.*5,0*.*2,0*.*…,0], respectively. As the defined correlation matrix (2) indicates variables in different groups that are uncorrelated, we also set the variance of each variable equal to 1, which makes the *Σ*_{x} a correlation matrix. We assume all variables in each group were equally relevant for prediction, and the covariances used in the simulation were $[{\sigma}_{x\stackrel{~}{y}1},\phantom{\rule{0.3em}{0ex}},{\sigma}_{x\stackrel{~}{y}14}]=[0.35,-0.3,0.25,0.2,0,\dots ,0]$, and simulated response vector was transformed to a binary variable, coded as −1 and 1. Only the four first groups of variables were relevant for classification due to their non-zero values of the covariance with *y*. Since variables are grouped, variable grouping information was used as background information, and ** Z** was coded as given in [15].

The data set is identical to that used in a recently conducted study [4], which did not utilize background information for parameter estimation. The data set contains 36 *Saccharomyces cerevisiae* strains, including the reference strain S288C [1,18]. Each strain has 16 chromosomes and a mitochondrial genome. In total 5791 protein-coding sequences, excluding dubious genes, were used as reference sequences. Each genome protein coding gene element, was converted into a vector of numeric features by sequence alignment of that element in a particular strain to the corresponding sequence element in the reference genome S288C. To achieve this, all reference sequences were first aligned against themselves, and for each reference sequence, the maximum alignment score, representing some coding gene of the S288C genome, was obtained. Then each individual *S. cerevisiae* genome was BLASTed against this reference set, using tblastx (http://blast.ncbi.nlm.nih.gov/Blast.cgi?CMD=Web&PAGE_TYPE=BlastHome). Hence for each genome sequence a maximum bit-score was obtained, providing a measure of to what extent the reference sequence was found in each individual genome. Since this score depends heavily on the length of the aligned sequences, numeric features were finally translated by Jukes-Cantor evolutionary distances. Numeric features were then assembled into a matrix *X*_{(N×K)} with *N*=36 rows and *K*=5791 columns, one column for each reference sequence element. Data for each phenotype was assembled into a column vector ** y** of length

The genome of the yeast reference strain S288C is exceptionally richly annotated on a functional level, reflected in that data on functional relatedness in terms of shared Gene Ontology (GO) annotations [21] exists for a vast majority of its gene products. In its most extreme form, this annotation denotes genes as essential or not essential for viability. Furthermore, information on strain specific genetic variations with a potentially large effect on phenotypes has recently been extracted from population genomic data [1,2]. This information takes the form of the presence or absence of specific gene amplifications, reflecting potentially phenotype changing gain-of-function mutations, and the presence or absence of premature stop codons and frameshifts, reflecting likely loss-of-function mutations with potential negative effect on phenotypes. These information elements, taken together, serve as background information classifying genotypes in the study. Background information was assembled into a matrix *Z*_{(L×K)} with *K*=5791 columns and *L*=51 rows, where each column represents a gene and each row represents a GO term (45) or a specific sequence variation (6). Each entry in ** Z**is denoted ‘1’ if the corresponding gene is associated with the respective GO term and denoted ‘0’ if not.

The data set consists of the column vector *y*_{(N×1)}representing each phenotype one at a time, the matrix *X*_{(N×K)}of genotypes based on Jukes-Cantor evolutionary distances and the matrix *Z*_{(L×K)} of background information on genes containing annotated levels. To mine for relations between phenotypes and genotypes, we implemented an L-PLS approach [15] utilizing the background information in the modeling stage. We employed a PLS based algorithm for parsimonious variable selection [17] which is implemented here for multivariate feature selection in two stages, first for selection of genotype variables ** X**and then for the selection of background information variables

The association between each phenotype vector ** y**and several genotype vectors

$$\begin{array}{ll}{y}_{0}& =y-{1}_{n}{\overline{y}}^{}\phantom{\rule{2em}{0ex}}& {X}_{00}& =X-{1}_{n}{\overline{x}}_{K}^{}\end{array}$$

(3)

in the following manner where $\overline{y}$ is the column mean vector of ** y**, ${\overline{x}}_{K}$, ${\overline{x}}_{N}$ and $\overline{\overline{x}}$ are the column-, row- and overall means of

Choose a value of *α*(0≤*α*≤1)For *a*=1,…,*A*components

Find latent v-vectors ${v}_{x2}^{a}$ (*K*×1) and ${v}_{z1}^{a}$ (*K*×1) by the NIPALS algorithm as shown in Figure Figure1,1, cycling through *y*_{a−1}, *X*_{a−1,a−1}and *Z*_{a−1}.

For chosen *α* set *w*_{x} as a linear combination of ${v}_{z1}^{a}$ and ${v}_{x2}^{a}$ and normalize to length equal to 1.

$$\begin{array}{lll}{w}_{x}^{a}& =\alpha {v}_{z1}^{a}+(1-\alpha ){v}_{x2}^{a}\phantom{\rule{2em}{0ex}}& \phantom{\rule{2em}{0ex}}\\ {w}_{x}^{a}& \leftarrow {w}_{x}^{a}/\left|\right|{w}_{x}^{a}\left|\right|\phantom{\rule{2em}{0ex}}& \phantom{\rule{2em}{0ex}}\end{array}$$

(4)

Construct score-vectors for ** X**and

$$\begin{array}{lll}{t}_{x}^{a}& ={X}_{a-1,a-1}{w}_{x}^{a}\phantom{\rule{2em}{0ex}}& \phantom{\rule{2em}{0ex}}\\ {t}_{z}^{a}& ={w}_{x}^{a}\phantom{\rule{2em}{0ex}}& \phantom{\rule{2em}{0ex}}\end{array}$$

(5)

Let ${T}_{x}=({t}_{x}^{1},\dots ,{t}_{x}^{a})$ and ${T}_{z}=({t}_{z}^{1},\dots ,{t}_{z}^{a})$ (=*W*_{x}). That is, the weights for ** X** are used as scores for

Compute ** Y**–,

$$\begin{array}{ll}{p}_{y}^{a}& ={y}_{a-1}^{}{p}_{x}^{a}& ={X}_{a-1,a-1}^{}{p}_{z}^{a}& ={Z}_{a-1}{t}_{z}^{a}{({{t}_{z}^{a}}^{}-1}^{\phantom{\rule{2em}{0ex}}}\phantom{\rule{2em}{0ex}}\end{array}$$

(6)

Let ${P}_{y}=({p}_{y}^{1},\dots ,{p}_{y}^{a})$, ${P}_{x}=({p}_{x}^{1},\dots ,{p}_{x}^{a})$ and ${P}_{z}=({p}_{z}^{1},\dots ,{p}_{z}^{a})$.

Deflate the data matrices to form residual matrices

$$\begin{array}{ll}{y}_{a}& ={y}_{a-1}-{t}_{x}^{a}{{p}_{y}^{a}}^{}\phantom{\rule{2em}{0ex}}& {X}_{a,a}& ={X}_{a-1,a-1}-{t}_{x}^{a}{{p}_{x}^{a}}^{}\phantom{\rule{2em}{0ex}}& {Z}_{a}& ={Z}_{a-1}-{p}_{z}^{a}{{t}_{z}^{a}}^{}\phantom{\rule{2em}{0ex}}\end{array}$$

(7)

end

In step 1) above *α*can be determined through cross-validation and the data driven choice of *α*indicates the relevancy of background information in modeling the genotype-phenotype relation. A large value of *α* indicates that background information, ** Z**, is highly relevant for explaining genotype-phenotype relations.

In essence, the L-PLS estimate of the regression coefficients for the above given model based on *A*components can be derived from the weights and loadings by:

$$\widehat{\beta}={W}_{x}{({P}_{x}^{}-1}^{{P}_{y}}$$

(8)

Selection of variables based on Variables Importance on Projection (VIP) [24] is an accepted approach in PLS. We [17] recently suggested a stepwise estimation algorithm for parsimonious variable selection, where a consistency based variable selection procedure is adopted, and data has been split randomly in a predefined number of subsets (test and training). For each split, a stepwise procedure is adopted to select the variables. Stable variables that are being selected by stepwise elimination from all split of data are selected finally. This algorithm was also implemented here, but multivariate feature selection was performed in two distinct stages, first for selection of genotype variables from ** X**and then for the selection of background information variables from

$$\phantom{\rule{-15.0pt}{0ex}}\mathrm{VI}{P}_{{X}_{k}}=\sqrt{K\sum _{a=1}^{A}\left[\left({{p}_{y}^{a}}^{},{\left({t}_{{x}_{k}}^{a}/{t}_{x}^{a},2\right)}^{},/,\sum _{a=1}^{A},\left({{p}_{y}^{a}}^{}\right)\right)\right]}$$

(9)

and

$$\phantom{\rule{-15.0pt}{0ex}}\mathrm{VI}{P}_{{Z}_{l}}=\sqrt{L\sum _{a=1}^{A}\left[\left({{p}_{y}^{a}}^{},{\left({t}_{{z}_{l}}^{a}/{t}_{z}^{a},2\right)}^{},/,\sum _{a=1}^{A},\left({{p}_{y}^{a}}^{}\right)\right)\right]}$$

(10)

The *VIP* weights the contribution of each variable according to the variance explained by each PLS component, and presents a combined effect of all components. Variable *k*can be eliminated if $\mathrm{VI}{P}_{{X}_{k}}<u$ and similarly variable *l* can be eliminated if $\mathrm{VI}{P}_{{Z}_{l}}<u$ for some user-defined threshold *u*[0,*∞*).

The stepwise elimination algorithm can be sketched as follows: Let *U*_{0}=** X**.

For iteration *g*run ** y**and

There are *M*criterion values below the cutoff *u*. If *M*=0, terminate the elimination here.

Else, let *S*=*fM* for some fraction *f*0,1]. Eliminate the variables corresponding to the *S*most extreme criterion values.

If there are still more than one variable left, let *U*_{g + 1}contain these variables, and return to 1).

The fraction *f*determines the ‘step length’ of the elimination algorithm, where an *f*close to 0 will only eliminate a few variables in every iteration. Elimination of variable in ** X**means an automatic elimination of the corresponding column variable in

From each iteration *g*of the elimination, we get a cross validated root mean square error (RMSE) from training data, here denoted by *E*_{g}. For both stages of the elimination, the number of influencing variables decreases at each iteration, and *E*_{g} will often decrease until some optimum is achieved, and then increase again as we keep on eliminating. A potentially much simpler model can be achieved by a relatively small increase in RMSE [17]. This means we need a rejection level *d*, where for each iteration beyond optimum root mean square error (RMSE) *E*^{} we can compute the t-test *p*-value between the optimum model response and the selected model response, to give a perspective on the trade-of between understandability of the model and the RMSE. Hence with a non-significant deviation from the optimum RMSE, a significant reduction in variables, and hence better understandability, can be achieved; for details, see [17].

Two variable selection methods, where background information was not used in modeling, were compared with our suggested procedure, where background information was used in the modeling step. Hence,in total three models were considered, one was our suggested L-PLS with 2-stage stepwise elimination of genes and background information (M1), the second was ordinary PLS with stepwise elimination [17] of genes only (M2), and the third was a Soft-Thresholding PLS (ST-PLS) [3] (M3). We have recently used the latter approach for mapping of genotype to phenotype information [4].

All methods were implemented in the R computing environment http://www.r-project.org/.

In the proposed algorithm it is possible to eliminate non-influential variables based on some threshold on *u* and also based on the statistical significance of *d*. Fixing the threshold *u*a priori could affect the performance of the algorithm. Thus, we performed variable elimination exclusively based on *d*, obtaining an optimum when using an upper limit of *u*=10. We also considered three step lengths (*f*=(0*.*1,0*.*5,1)). In the first regularization step, we tried different rejection levels (*d*=(0*.*80,0*.*90,0*.*95,0*.*99); hence, in each iteration, variable elimination was based on statistical significance. Model performance was computed in the L-PLS fitting and regularized elimination fitting stages. For accurate model estimation and to avoid over fitting, the data was split at three levels. Figure Figure33 gives a graphical overview of the procedure. At level 1, we split the data into a test set containing 25% of the genomes and a training set containing the remaining 75%. This split was repeated 10 times, each time sampling the test set at random, i.e. the 10 test and training sets were partially overlapping. These splits were used for model evaluation, where in each of the 10 instances, selected variables were used for classifying the level 1 test set, and RMSE was computed. In the regularized elimination fitting, there are two levels of cross-validation, as indicated by the right section of Figure Figure3.3. First, a 10-fold cross-validation was used to optimize the fraction *f* and the rejection level *d*in the elimination part of our algorithm. Second, at the final stage, leave-one-out cross-validation was used to estimate all parameters in the L-PLS method. These two procedures together corresponds to a comprehensive ‘cross-model validation’ [17,25]. Note that the above split of the data was done on ** y** and

For the simulation data, a power analysis was conducted and results are presented in Figure Figure4.4. Here, the power of selecting the correct variables as function of the information content of ** Z** in L-PLS

To study the impact of background information on explanatory variables for genotype-phenotype relations in yeast, a 2-stage stepwise backward elimination procedure in L-PLS was used. We modelled each phenotype separately. The algorithm was illustrated using Melibiose Rate as an example. However, the performance was very similar also for other responses (phenotypes), as presented in Table Table1.1. In total, we fitted 20 models, one for each phenotype. First, a genotype predictor matrix was derived by blasting the genes of each genome to a *S. cerevisiae* reference genome, and the best hit scores were used as numerical inputs to a genotype matrix [4]. Gene ontology terms, reflecting functional relatedness with regards to the gene product participation in similar molecular processes, together with data on gene dispensability (essential/not essential) and data on the number of gene paralogs present in founder genome, were used as background information. This data essentially reflects gene relationships in the S288C reference genome; relationships which may or may not be conserved in the species as a whole. We also included population genomic data reflecting the presence or absence, in each specific strain, of genetic variations with a potentially large impact on phenotypes. Gene copy number variations reflect potential gain-of-function mutations in particular lineages, whereas frameshift and premature stop codon mutations reflect potential loss-of-function mutations in respective lineages. The proposed model was fitted to each of the 20 phenotypes, and results are summarized in Table Table1.1. Figure Figure55 exemplifies the progression of the 2-stage variable elimination for the phenotype Melibiose 2% Rate, representing the rate of growth of the set of yeast strains when supplied with carbon exclusively in the form of the melibiose. The number of genotype variables, ** X**, and background information,

A key requirement of any multivariate analysis is the stability and selectivity of the results. To evaluate model stability and selectivity, we [17] recently introduced a simple *selectivity score*: if a variable is selected as one out of *m* variables, it will get a score of 1/*m*. Repeating the selection for each split of the data, we simply add up the scores for each variable. Thus, a variable having a large selectivity score tends to be repeatedly selected as one among a few variables. In Figure Figure7,7, the selectivity score is sorted in descending order and is presented for X-variables (genes) in the upper left panel for M1, the upper right panel for M2 and the lower left panel for M3. The selectivity score indicating the stability of the selected Z-variables (GO- terms) obtained from M1 is presented in the lower left panel. M1 indeed selected many genes in a stable way, which is a fundamental requirement for any further analysis. A selectivity score above 0.2 for X-variables and above 0.06 for Z-variables is significantly larger than similar scores obtained by repeated fitting of models using random permutation on the phenotypes. Since traits are controlled by subsets of distinct genes [4], and some genes in the genome are of overall importance for handling variations in the external environment and affect a disproportionate number of phenotypes [26], we expect any method extracting relevant biological information to have a higher selectivity score than any random selection of genes. This was indeed the case for our proposed method M1. In fact, using the two-step L-PLS procedure, only 30 genes were selected from M1, corresponding to substantially higher selectivity than M2 and M3. Not surprisingly, these genes OLI1, YEH1, ATP8, PSY3, IFM1, SUV3, CAR1, ERG6, ILS1, YDR374C, SHO1, YDR476C, GLO3, APL5, RIX1, GPR1, VAR1, TTI2, YLR410WB, YDL211C, YDL218W, EHD3, MRPL28, RPT6, COX17, STE11, SUR4, YAP1, MRPL39, YNL320W, were involved in cellular functions directly relating to variations in the environment: transport, stress response, response to chemical stimulus and metabolism. They also tended to be affected by both strong loss-of-function (premature stop codons, frameshifts) and gain-of-function (copy number variation) mutations, as presented in Table Table2.2. We found 72.% genes overlap between M1 and M2 and 67.9% genes overlap between M1 and M3 for Melibiose 2% Rate. The selection of background variables can be missed if only a few of the corresponding genes are significant [14], but the powerful structure of L-PLS, coupled with 2-stage stepwise elimination procedure, yields a stable list of genes and background information variables, which maps the genotype-phenotype relation. Finally, we have listed the mapped background information and genes for all 20 phenotypes in Table Table11 and Additional file 1: Table S1 respectively.

Assuming that phenotypic variation within the species is controlled by either or both of lineage specific adaptive mutations, emerging as a consequence of lineage specific positive selection, or neutral variation, emerging as consequence of lineage specific relaxation of selective pressure that allow loss-of-function mutations to accumulate, we expected phenotype defining genes to show faster evolution than non-influential genes. This corresponds to a prediction of a higher ratio of nonsynonymous versus synonymous mutations since the split between *S. cerevisiae* and its closest relative *Saccharomyces paradoxus*[27]. Indeed, we found genes identified as influential through M1 to have been evolving 29% faster than non-influential genes (*p*<0*.*10). This indicates that these genes, as a group, have been subjected to either stronger positive selection or somewhat relaxed negative selection during the recent yeast history and supports that M1 extracts biologically relevant information.

We have suggested the use of background information in the modeling step for genotype-phenotype mapping through L-PLS and a stepwise elimination procedure. We note that the derived results could give a slight decrease in RMSEP when background information is used, but more interestingly, this comes with more stability in the selection of variables (genes, GO terms and variations) used for genotype-phenotype mapping. We conclude that the approach is worth pursuing, and future investigations should be made to improve the computations of genotype signals, and variable selection procedure within the PLS framework.

The authors declare that they have no competing interests.

TM initiated the project and the ideas. SS has been involved in the later development of the approach and the final algorithm. TM has done the programming and computations, with assistance from SS. TM has drafted the manuscript, with inputs SS, JW and LS. All authors have read and approved the final manuscript.

**Table S1.** Selectivity score based selected genes. Genes selected for each phenotype for genotype phenotype mapping by using 2-stage variable elimination and having selectivity score above 0.06.

Click here for file^{(18K, pdf)}

Tahir Mehmoods scholarship has been fully financed by the Higher Education Commission of Pakistan, Jonas Warringer was supported by grants from the Royal Swedish Academy of Sciences and Carl Trygger Foundation.

- Liti G, Carter DM, Moses AM, Warringer J, Parts L, James SA, Davey RP, Roberts IN, Burt A, Koufopanou V, Tsai IJ, Bergman CM, Bensasson D, O’Kelly MJT, van Oudenaarden A, Barton DBH, Bailes E, Nguyen AN, Jones M, Quail MA, Goodhead I, Sims S, Smith F, Blomberg A, Durbin R, Louis EJ. Population genomics of domestic and wild yeasts. Nature. 2009;458:337–341. doi: 10.1038/nature07743. [PMC free article] [PubMed] [Cross Ref]
- Warringer J, Zörgö E, Cubillos F, Zia A, Gjuvsland A, Simpson J, Forsmark A, Durbin R, Omholt S, Louis E. et al. Trait variation in yeast is defined by population history. PLoS Genet. 2011;7(6):e1002111. doi: 10.1371/journal.pgen.1002111. [PMC free article] [PubMed] [Cross Ref]
- Sæbø S, Almøy T, Aarøe J. Aastveit AH. ST-PLS: a multi-dimensional nearest shrunken centroid type classifier via PLS. J Chemometrics. 2007;20:54–62.
- Mehmood T, Martens H, Saebo S, Warringer J, Snipen L. Mining for genotype-phenotype relations in saccharomyces using partial least squares. BMC Bioinformatics. 2011;12:318. doi: 10.1186/1471-2105-12-318. [PMC free article] [PubMed] [Cross Ref]
- Badano A. Modeling the bidirectional reflectance of emissive displays. Appl Opt. 2002;41:3847–3852. doi: 10.1364/AO.41.003847. [PubMed] [Cross Ref]
- Allison DB, Thiel B, Jean PS, Elston RC, Infante MC, Schork NJ. Multiple phenotype modeling in gene-mapping studies of quantitative traits: power advantages. Am J Hum Genet. 1998;63:1190–1201. doi: 10.1086/302038. [PubMed] [Cross Ref]
- Kraft P. de Andrade M. Group 6: Pleiotropy and multivariate analysis. Genet Epidemiol. 2003;25(Suppl 1):S50—S56. [PubMed]
- Giaever G, Flaherty P, Kumm J, Proctor M, Nislow C, Jaramillo DF, Chu AM, Jordan MI, Arkin AP, Davis RW. Chemogenomic profiling: identifying the functional interactions of small molecules in yeast. Nat Acad Sci. 2004;101:793–798. doi: 10.1073/pnas.0307490100. [PubMed] [Cross Ref]
- Parsons AB, Lopez A, Givoni IE, Williams DE, Gray CA, Porter J, Chua G, Sopko R, Brost RL, Ho CH, Wang J, Ketela T, Brenner C, Brill JA, Fernandez GE, Lorenz TC, Payne GS, Ishihara S, Ohya Y, Andrews B, Hughes TR, Frey BJ, Graham TR, Andersen RJ. Boone C. Exploring the mode-of-action of bioactive compounds by chemical-genetic profiling in yeast. Cell. 2006;126:611–625. doi: 10.1016/j.cell.2006.06.040. [PubMed] [Cross Ref]
- Chun H, Keleş S. Sparse partial least squares regression for simultaneous dimension reduction and variable selection. J R Stat Soc: Ser B (Statistical Methodology) 2010;72:3–25. doi: 10.1111/j.1467-9868.2009.00723.x. [PMC free article] [PubMed] [Cross Ref]
- Wende A, Huss J, Schaeffer P, Giguere V, Kelly D. PGC-1 coactivates PDK4 gene expression via the orphan nuclear receptor ERR: a mechanism for transcriptional control of muscle glucose metabolism. Mol Cell Biol. 2005;25:10684–10694. doi: 10.1128/MCB.25.24.10684-10694.2005. [PMC free article] [PubMed] [Cross Ref]
- Jorgensen K, Hjelle S, Oye O, Puntervoll P, Reikvam H, Skavland J, Anderssen E, Bruserud O, Gjertsen B. Untangling the intracellular signalling network in cancer-a strategy for data integration in acute myeloid leukaemia. J Proteomics. 2011;74(3):269–281. doi: 10.1016/j.jprot.2010.11.003. [PubMed] [Cross Ref]
- Mootha V, Lindgren C, Eriksson K, Subramanian A, Sihag S, Lehar J, Puigserver P, Carlsson E, Ridderstrale M, Laurila E. et al. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet. 2003;34(3):267–273. doi: 10.1038/ng1180. [PubMed] [Cross Ref]
- Tomfohr J, Lu J, Kepler T. Pathway level analysis of gene expression using singular value decomposition. BMC Bioinformatics. 2005;6:225. doi: 10.1186/1471-2105-6-225. [PMC free article] [PubMed] [Cross Ref]
- Sæbø S, Almøy T, Flatberg A, Aastveit A, Martens H. LPLS-regression: a method for prediction and classification under the influence of background information on predictor variables. Chemometrics Intell Lab Syst. 2008;91(2):121–132. doi: 10.1016/j.chemolab.2007.10.006. [Cross Ref]
- Vinzi V, Chin W, Henseler J. Handbook of Partial Least Squares: Concepts, Methods and Applications. Springer; 2010.
- Mehmood T, Martens H, Sæbø S, Warringer J, Snipen L. A partial least squares based algorithm for parsimonious variable selection. Algorithms Mol Biol. 2011;6:27–38. doi: 10.1186/1748-7188-6-27. [PMC free article] [PubMed] [Cross Ref]
- Liti G, Louis EJ. Yeast evolution and comparative genomics. Annu Rev Microbiol. 2005;59:135–153. doi: 10.1146/annurev.micro.59.030804.121400. [PubMed] [Cross Ref]
- Warringer J, Zorgo E, Cubillos FA, Gjuvsland A, Louis EJ, Omholt S, Liti G, Moses A, Blomberg A. Trait variation in yeast is defined by population history. PLoS Genet. 2011;7:1–15. [PMC free article] [PubMed]
- Warringer J, Anevski D, Liu B, Blomberg A. Chemogenetic fingerprinting by analysis of cellular growth dynamics. BMC Chem Biol. 2008;8:3. doi: 10.1186/1472-6769-8-3. [PMC free article] [PubMed] [Cross Ref]
- Dimmer EC, Huntley RP, Barrell DG, Binns D, Draghici S, Camon EB, Hubank M, Talmud PJ, Apweiler R, Lovering RC. The gene ontology - providing a functional role in proteomic studies. Proteomics. 2008;8
- Martens H, Anderssen E, Flatberg A, Gidskehaug L, Høy M, Westad F, Thybo A, Martens M. Regression of a data matrix on descriptors of both its rows and of its columns via latent variables: L-PLSR. Comput Stat Data Anal. 2005;48:103–123. doi: 10.1016/j.csda.2003.10.004. [Cross Ref]
- Eriksson L, Damborsky J, Earll M, Johansson E, Trygg J, Wold S. Three-block bi-focal PLS (3BIF-PLS) and its application in QSAR. SAR QSAR Environ Res. 2004;15(5-6):481–499. doi: 10.1080/10629360412331297452. [PubMed] [Cross Ref]
- Nemeth M. Multi-and megavariate data analysis. Technometrics. 2003;45(4):362–362.
- Anderssen E, Dyrstad K, Westad F, Martens H. Reducing over-optimism in variable selection by cross-model validation. Chemometrics Intell Lab Syst. 2006;84(1-2):69–74. doi: 10.1016/j.chemolab.2006.04.021. [Cross Ref]
- Hillenmeyer M. PhD thesis. USA: Stanford University; 2009. Identifying relationships between genes and small molecules, from yeast to humans.
- Wall D, Hirsh A, Fraser H, Kumm J, Giaever G, Eisen M, Feldman M. Functional genomic analysis of the rates of protein evolution. Proc Nat Acad Sci USA. 2005;102(15):5483. doi: 10.1073/pnas.0501761102. [PubMed] [Cross Ref]

Articles from BMC Bioinformatics are provided here courtesy of **BioMed Central**

PubMed Central Canada is a service of the Canadian Institutes of Health Research (CIHR) working in partnership with the National Research Council's Canada Institute for Scientific and Technical Information in cooperation with the National Center for Biotechnology Information at the U.S. National Library of Medicine(NCBI/NLM). It includes content provided to the PubMed Central International archive by participating publishers. |