You are on page 1of 12

EurekaBrewing.wordpress.com September 2015

Insight into the Dekkera anomala YV396 genome

Samuel H. Aeschlimann*

*EurekaBrewing.wordpress.com (powered by Blackwell Brewery)

Abstract

The yeast Dekkera (anamorph Brettanomyces) is associated with various industrial fermentations and currently experiences an increase of interest in the Wild Ale beer brewing scene. The notorious spoilage yeast is capable of forming volatile phenols with flavor descripties like ’wet animal, musty’. Due to the high abundance of spoilage in the wine industry, so far only genomes of Brettanomyces bruxellensis have been published and discussed. Other species from the Dekkera taxa include Dekkera anomala mainly associated with beers such as Belgian Lambics. The current draft Dekkera anomala YV396 genome assembly consists of 30 contigs with an estimated genome size of 12.9 Mb. Addressing the degree of completeness of the genome revealed the existance of a set of tRNAs capable of encoding the standard 20 amino acids including two tRNAs encoding for selenocysteine. Furthermore the entire set of 248 CEGs could be found in the assembly indicating a rather complete draft genome. 4,000 genes were predicted of which the main part has homologues in Brettanomyces bruxellensis. Two phenolic acid decarboxylses as well as one vinylphenol reductase - two enzymes possibly associated with volatile phenols - could be further identified.

The genome as well as the proteome provide a useful resource for additional analysis of pathways associates with flavor compounds in beer production.

EurekaBrewing.wordpress.com • September 2015 Insight into the Dekkera anomala YV396 genome S amuel H. A eschlimann

Figure 1: Micrograph of a Dekkera anomala strain (EYL012 Brettanomyces cantillon VI)

EurekaBrewing.wordpress.com September 2015

Contents

I

Introduction

 

2

II

Results and Discussion

3

I

Genome assembly

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

3

II

tRNA analysis . .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

4

III

IV

Gene prediction

. Gene annotation . . . . . .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

4

4

V

Carbohydrate metabolism

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

6

VI

Phenolic acid metabolism

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

7

 

VI.1

Phenolic acid decarboxylase

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

7

VI.2

Vinylphenol reductase

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

9

III Methods

 

10

 

I

Genome assembly

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

10

II

Gene prediction

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

10

III

Gene annotation .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

10

 

I.

Introduction

 

Dekkera (anamorph Brettanomyces) are notorious beverage spoilage yeasts associated with a variety of fermentation products such as wine, beer, cider, kombucha as well as sourdough [Yakobson, 2010; White and Zainasheff, 2010]. Due to its spoilage potential and impact on wine quality, a lot of research has been conducted on Dekkera to further understand the associated pathways leading to the unwanted off-flavors. Volatile phenols (4-ethylphenol, 4-ethylguaiacol and 4-ethylcatechol) with sensory descriptives like ’animal, musty’ are especially in the focus of the wine industry as these compounds exhibit low detection thresholds [Curtin et al., 2012]. Beside its importance in the wine industry, Dekkera is also present in Belgian beers like Lambics and Gueuzes where the volatile phenols created by Dekkera species are all part of the desired beer profile [Spitaels et al.,

2014].

Next generation sequencing enables us to look at the genomic setup of organisms to better understand the organisms and metabolic pathways. This publication discusses the genome assembly created by Vervoort et al. [2015], the gene prediction as well as certain pathways linked to beer production.

EurekaBrewing.wordpress.com September 2015

  • I. Genome assembly

II.

Results and Discussion

The current D. anomala genome consists of 30 contigs with an estimated genome size of about 12.9 Mb (Tab 1). The estimated genome is similar in size to the close relative D. bruxellensis or Saccharomyces cerevisiae [Goffeau et al., 1996; Curtin et al., 2012]. Looking at the GC content of the D. anomala draft genome in comparison to other Dekkera genomes reveales a similar distribution pattern (Fig 2).

Table 1: D. anomala YV396 genome statistics (accession number GCA_001005505.1)

Name

Total sequence length [bp] Number of contigs [n] Contig N50 [bp] 12,880,761 30 1,112,461
Total sequence length [bp]
Number of contigs [n]
Contig N50 [bp]
12,880,761
30
1,112,461

Figure 2: GC plots for D. bruxellensis CBS2499 [Piškur et al., 2012], D. bruxellensis AWRI1499 [Curtin et al., 2012], D. bruxellensis LAMAP 2480 [Valdes et al., 2014] and D. anomala YV396 [Vervoort et al., 2015]

EurekaBrewing.wordpress.com September 2015

II.

tRNA analysis

To assess the completeness of the genome assembly in terms of tRNAs, tRNAscan-SE (Vers 1.3.1, default settings) established by Lowe and Eddy, 1997 was run on the genome to predict tRNAs. tRNAscan-SE predicted 123 tRNAs in the D. anomala genome and a complete set of tRNAs sufficient to encode the standard 20 amino acids including two tRNAs encoding for selenocysteine (not shown). The real number of tRNAs was not addressed since the tRNAs were only used as a parameter to address the completeness of the genome assembly.

III.

Gene prediction

The results of the Augustus gene prediction, as further explained in chapter II, are summarized in Tab 2. 4,050 potential genes and 4,160 potential transcripts could be predicted. The average coding sequence length is about 1,600 bp with the longest transcript length of 14,904 bp (Fig 3a). Exons are generally shortern than 4,000 bp (Fig 3b) whereas introns are commonly shorter than 1,000 bp (Fig 3c). A gene in the genome of D. anomala contains on average 1.3 exons and 0.3 introns.

The shortest protein (67 aa) encodes a 40S ribosomal protein s28-a (part of the small ribosomal subunit) whereas the longest 4,967 encodes for midasin (nuclear chaperone). RPS28A, the 40S ribosomal protein homologue in S. cerevisiae has a length of 67 aa as well (Uniprot). MDN1, the midasin homologue in S. cerevisiae has a length of 4,910 aa (Uniprot). The average protein length distribution is shown in Fig 3d).

IV.

Gene annotation

Blast2GO annotated about 3,000 out of the 4,160 predicted genes (Fig 8). Woolfit et al., 2007 reported about 3,000 genes for a D. bruxellensis assembly, Curtin et al., 2012 about 4,969 and Piškur et al., 2012 of about 5,600 genes.

To address the completeness of the assembly in terms of proteins, a set of 248 highly conserved core proteins deposited by Parra et al., 2009 was searched for in the predicted protein sequences. A blastp analysis revealed hits for 219 CEGs (core eukaryotic genes) with a coverage greater than 70% and e-value threshold of 1e-10. A tblastn analsyis further revealed hits for 28 of the remaining CEGs. The remaining CEG candidate (KOG3479 Mitochondrial import inner membrane translocase, subunit TIM9 [Intracellular trafficking, secretion, and vesicular transport]) could be found in the initial blastp results (hit = coverage of 69% and a sequence identity of 92%) but could not be picked up by the blast-result parsing scripts due to the sequence coverage threshold set to 70%. The hereby presented draft genome assembly of D. anomala therefore very likely contains the entire set of 248 CEG proteins.

All the investigations to address the completeness of the genome indicate a rather complete genome assembly of D. anomala strain YV396. The next chapters discuss certain pathways and enzymes of importance to beer production.

EurekaBrewing.wordpress.com September 2015

EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron

EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron

(a) Transcript sequence length distribution

(b) Exon sequence length distribution

EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron
EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron
EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron
EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron

EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron
EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron
EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron
EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron
EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron
EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron
EurekaBrewing.wordpress.com • September 2015 (a) Transcript sequence length distribution (b) Exon sequence length distribution (c) Intron

(c) Intron sequence length distribution

(d) Protein sequence length distribution

Figure 3: Length distribution histograms for D. anomala transcripts, exons, introns as well as proteins

Table 2: D. anomala YV396 gene prediction statistics

Name

Total predicted genes

4,050

Average gene length [bp]

1,715

Total predicted transcripts

4,160

Average transcript length

1,742

Introns [n]

1,277

Average intron length [bp]

473

Exons [n]

5,435

Average exon length [bp]

1,221

EurekaBrewing.wordpress.com September 2015

  • V. Carbohydrate metabolism

Various enzymes are part of the carbohydrate metabolism in yeasts (Fig 4). The most important ones for beer production are glucosidases cleaving sugars from polysaccharides such as maltose, sucrose or even starch and cellulose.

At least one alpha-glucosidase (EC 3.2.1.20) could be predicted in the D. anomala genome which was however not annotated by the Blast2GO pipeline (not colorized EC 3.2.1.20 in Fig 4). Alpha- glucosidase is capable of cleaving terminal 1>4 linked sugars of higher sugars such as sucrose and starch. The predicted D. anomala alpha-glucosidase is 949 amino acids in length and shows high sequence identities to alpha-glucosidase found in other yeasts like Candida albicans.

One beta-glucosidase candidate protein (also called cellulase, EC 3.2.1.21) could be predicted in the D. anomala genome. This enzyme can cleave 1>4 beta-D-glucose from polysaccharides like cellulose. It is 841 amino acids in length and shows high sequence identity to a D. bruxellensis beta-glucosidase (90% sequence identity, total length 841 amino acids, EIF45415.1, deposited by Curtin et al., 2012). The sequence of the predicted D. anomala beta-glucosidase from this genome assembly was already deposited on NCBI by Vervoort et al., 2015 with the accession number

AKS48904.1.

EurekaBrewing.wordpress.com • September 2015 V. Carbohydrate metabolism Various enzymes are part of the carbohydrate metabolism in

Figure 4: Enzymes linked to starch and sucrose metabolism - Enzymes highlighted in color could be predicted and annotated on the D. anomala genome

EurekaBrewing.wordpress.com September 2015

VI.

Phenolic acid metabolism

The spoilage potential of Brettanomyces is mainly associated with volatile phenols like 4-ethylphenol (animal, leather, horse sweat) and 4-ethylguaiacol (spicy, clove-like) [Doss, 2008]. These compounds have rather low sensory threshold levels and can easily be picked up in products like wine [Campolongo et al., 2010]. Two key enzymes are involved in the metabolism of hydroxycinnamic acids into volatile phenols (Fig 5). The hydroxycinnamic acids originate from cinnamic acids which are metabolized by an esterase [Kheir et al., 2013]. Cinnamic acids are part of plant cell walls and are associated with antimicrobial function Godoy et al., 2009.

The first enzyme, a phenolic acid decarboxylase (padc), is responsible for the conversion of hydroxycinnamic acids into vinyl derivatives. Another enzyme (vinylphenol reductase (vpr)) then reduces the vinyl derivatives into ethyl phenols. The following two chapters discuss PADC and VPR in more detail.

EurekaBrewing.wordpress.com • September 2015 VI. Phenolic acid metabolism The spoilage potential of Brettanomyces is mainly associated

Figure 5: Phenolic acid metabolism - based on two enzymes. First row of compounds (top-bottom) represent hydrox- ycinnamic acids, second row represent vinyl derivaties, third row represent ethyl derivatives

VI.1 Phenolic acid decarboxylase

Edlin et al., 1995 first identified and described the function of a D. anomala enzyme capable of transforming p-coumaric acids into 4-vinyl derivaties: phenolic acid decarboxylase (padc). PADC have homologues in other species such as S. cerevisiae, Lactobacillus plantarum, Pediococcus pentosaceus or Pseudomonas fluorescens [Harris et al., 2009]. The first sequencing projects of D. bruxellensis by Woolfit et al., 2007 revealed a protein with sequence similarities to Protoplast secreted protein 2 (PST2) found in S. cerevisiae. PST2 is however not linked to any hydroxycinnamic acid metabolism up to today.

EurekaBrewing.wordpress.com September 2015

EurekaBrewing.wordpress.com • September 2015 (a) Phylogenetic tree (b) PADC homologue structure (c) PST2 homologue structure Figure

(a) Phylogenetic tree

EurekaBrewing.wordpress.com • September 2015 (a) Phylogenetic tree (b) PADC homologue structure (c) PST2 homologue structure Figure

(b) PADC homologue structure

EurekaBrewing.wordpress.com • September 2015 (a) Phylogenetic tree (b) PADC homologue structure (c) PST2 homologue structure Figure

(c) PST2 homologue structure

Figure 6: (a) Phylogenetic analysis of PADC homologues from different species - analysis performed using http://www.phylogeny.fr (b) SWISS-MODEL for PADC homologue (c) SWISS-MODEL for PST2 homologue

The D. anomala draft genome contains two different sequences that can be linked to either PST2 or PADC (Fig 6a). The PADC homologue is 176 amino acids in length with close sequence identities to a predicted protein from a fungi called Nectria haematococca (XP_003042417) and contains a phenolic acid decarboxylase domain (IPR008729). The PST2 homologue is around 258 amino acids in length with close sequence identities to B. bruxellensis protoplast secreted protein 2 precursor, EIF46519, and contains a flavodoxin binding site. These results suggest the presence of two proteins that might be involved in the metabolism of hydroxycinnamic acids into vinyl derivatives.

To further evaluate the possible function of the two proteins, a protein structure prediction was performed using SWISSPROT [Biasini et al., 2014; Arnold et al., 2006]. The D. anomala PADC homologue revealed a structre modeled on a ferulic acid decarboxylase model working as homo- dimers (Fig 6b). Indicating a possible function as a decarboxylase like PADC. The structure of the PST2 homologue was modeled based on a flavoprotein wrbA and resulted in a homo-tetramer structure (Fig 6c). Further investigations are necessary to understand and address whether these

EurekaBrewing.wordpress.com September 2015

two phenolic acid decarboxylase protein candidates are actually linked to the phenolic acid metabolism.

VI.2 Vinylphenol reductase

Vinylphenol reductase (VPR) is a very unique enzyme present only in yeasts like Dekkera [Tchobanov et al., 2008]. The enzyme can turn vinyl derivatives such as 4-vinyl guaiacol, 4-vinylphenol or 4- vinylcatechol into compounds like 4-ethyl guaiacol, 4-ethyl phenol or 4-ethyl catechol respectively (Fig 5).

Tchobanov et al., 2008 were of the first to characterize VPR. The authors purified VPR from D. bruxellensis and charazterized the enzyme in more detail. The empirically determined protein weight was around 26 kDa (SDS-PAGE) and a length of 210 amino acids was determined by peptide sequencing. The authors then generated an in silico DNA sequence from the peptide sequence. Parts of the DNA sequence were later on confirmed by Campolongo et al., 2010.

The draft genome of Dekkera anomala contains one predicted protein sequence with high sequence similarity to the in silico sequence provided by Tchobanov et al., 2008. The retreived sequence is 210 amino acids in length and shows about 77.6% sequence identity compared to the predicted in-silico sequence (Fig 7). Although the VPR activity is reported to be NADH dependent [Kheir et al., 2013], no NADH binding site could be predicted for both sequences.

EurekaBrewing.wordpress.com • September 2015 two phenolic acid decarboxylase protein candidates are actually linked to the phenolic

Figure 7: Sequence alignment of predicted VPR protein from D. anomala (top) versus the predicted in-silico sequence from Tchobanov et al., 2008 (bottom)

EurekaBrewing.wordpress.com September 2015

  • I. Genome assembly

III.

Methods

The draft genome assembly of Dekkera anomala strain YV396 (isolated from a Belgian brewery) was retrieved from GenBank (accession number LCTY00000000.1; June 2015) deposited in May 2015 by KU Leuven Vervoort et al. [2015]. Illumina HiSeq data (100x coverage) was assembled into a genome using SOAPdenovo v.1.05. The statistics for the obtained assembly are summarized in Tab. 1).

II.

Gene prediction

Gene prediction on contigs was performed using the AUGUSTUS web-service (AUGUSTUS parameter project identifier: pichia_stipitis, UTR prediction: false, report genes on both strands, alternative transcripts few, allowed gene structure: predict any number of (possibly partial) genes, ignore conflicts with other strand: false) [Stanke et al., 2006, 2008]. The gene prediction statistics are summarized in Tab. 2.

III.

Gene annotation

Gene annotation was performed by Blast2GO including remote blastx on NCBI and InterProScan for domain predictions [Conesa et al., 2005]. GO-term mapping and annotation performed by Blast2GO pipeline. Close to 3,000 out of the predicted 4,160 sequences could be annotated by Blast2GO (Fig 8). Another subset of about 600 sequences could be mapped to a biological function without a GO term and about 460 sequences only resulted in BLAST hits which could not be further associated with a protein function.

Most abundant species associated with the best blastx hits were Dekkera bruxellensis, Ogataea polymorpha and Pichia kudriavzevi (not shown).

EurekaBrewing.wordpress.com • September 2015 I. Genome assembly III. M ethods The draft genome assembly of Dekkera

Figure 8: Distribution of predicted D. anomala genes after Blast2GO annotation

EurekaBrewing.wordpress.com September 2015

References

Arnold, K., Bordoli, L., Kopp, J., and Schwede, T. (2006). The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics, 22:195–201.

Biasini, M., Bienert, S., Waterhouse, A., Arnold, K., Studer, G., Schmidt, T., Kiefer, F., Cassarino, T., Bordoli, L., and Schwede, T. (2014). SWISS-MODEL: modelling protein tertiary and quaternary structure using evolutionary information. Nucleic Acids Research, 42.

Campolongo, S., Lamberti, C., Rantsiou, K., Pessione, E., and Cocolin, L. (2009-2010). Brettanomyces bruxellensis and off-odours: Genetic and proteomic approaches to unravel the molecular mecha- nism of ethyl-phenol production. QUAD. VITIC. ENOL. UNIV. TORINO, 31.

Conesa, A., Götz, S., García-Gómez, J. M., Terol, J., Talón, M., and Robles, M. (2005). Blast2GO:

a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics, 21(18):3674–3676.

Curtin, C., Borneman, A., Chambers, P., and Pretorius, I. (2012). De-novo assembly and analysis of the heterozygous triploid genome of the wine spoilage yeast Dekkera bruxellensis AWRI1499. PLoS One, 7(3).

Doss, G. (2008). Brettanomyces: Flavors and performance of single and multiple strain fermentations with respect to time. AHA NHC June 2008 lecture slides.

Edlin, D., Narbad, A., J.R., D., and Lloyd, D. (1995). The biotransformation of simple phenolic compounds by Brettanomyces anomalus. FEMS Microbiology Letters, 125:311–316.

Godoy, L., Garrido, D., Martinez, C., Saavedra, J., Combina, M., and Ganga, M. (2009). Study of the coumarate decarboxylase and vinylphenol reductase activities of Dekkera bruxellensis (anamorph Brettanomyces bruxellensis) isolates. Letters in Applied Microbiology, 48:452–4557.

Goffeau, A., Barrell, B. G., Bussey, H., Davis, R. W., Dujon, B., Feldmann, H., Galibert, F., Hoheisel, J. D., Jacq, C., Johnston, M., Louis, E. J., Mewes, H. W., Murakami, Y., Philippsen, P., Tettelin, H., and Oliver, S. G. (1996). Life with 6000 Genes. Science, 274(5287):546–567.

Harris, V., Ford, C., Jiranek, V., and Grbin, P. (2009). Survey of enzyme activity responsible for phenolic off-flavour production by Dekkera and Brettanomyces yeast. Appl Microbiol Biotechnol,

81:1117–1127.

Kheir, J., Salameh, D., Strehaiano, P., Brandman, C., and Lteif, R. (2013). Impact of volatile phenols and their precursors on wine quality and control measures of Brettanomyces/Dekkera yeasts. Eur Food Res Technol.

Lowe, T. M. and Eddy, S. R. (1997). tRNAscan-SE: A Program for Improved Detection of Transfer RNA Genes in Genomic Sequence. Nucleic Acids Research, 25(5):0955–964.

Parra, G., Bradnam, K., Ning, Z., Keane, T., and Korf, I. (2009). Assessing the gene space in draft genomes. Nucleic Acids Research, 37(1):289–297.

Piškur, J., Ling, Z., Marcet-Houben, M., Ishchuk, O., Aerts, A., LaButti, K., Copeland, A., Lindquist, E., Barry, K., Compagno, C., Bisson, L., Grigoriev, I., Gabaldón, T., and Phister, T. (2012). The genome of wine yeast Dekkera bruxellensis provides a tool to explore its food-related properties. International Journal of Food Microbiology, 157:202–209.

EurekaBrewing.wordpress.com September 2015

Spitaels, F., D., W. A., Maarten, J., Maarten, A., Heide-Marie, D., Anita, V. L., Luc, D. V., and Peter, V. (2014). The Microbial Diversity of Traditional Spontaneously Fermented Lambic Beer. PLoS ONE, 9(4):e95384.

Stanke, M., Diekhans, M., Baertsch, R., and Haussler, D. (2008). Using native and syntenically mapped cDNA alignments to improve de novo gene finding. Bioinformatics, 24(5):637–644.

Stanke, M., Schoffmann, O., Morgenstern, B., and Waack, S. (2006). Gene prediction in eukary- otes with a generalized hidden Markov model that uses hints from external sources. BMC Bioinformatics, 7(1):62.

Tchobanov, I., Gal, L., Guilloux-Benatier, M., Remize, F., Nardi, T., Guzzo, J., Serpaggi, V., and Alexandre, H. (2008). Partial vinylphenol reductase purification and characterization from Brettanomyces bruxellensis. FEMS Microbiol. Lett., 284:213–217.

Valdes, J., Tapia, P., Cepeda, V., Varela, J., Godoy, L., Cubillos, F., Silva, E., Martinez, C., and Ganga, M. (2014). Draft genome sequence and transcriptome analysis of the wine spoilage yeast Dekkera bruxellensis LAMAP2480 provides insights into genetic diversity, metabolism and survival. FEMS Microbiology Letters, 361(2).

Vervoort, Y., Herrera-Malaver, B., Mertens, S., Guadalupe Medina, V., Duitama, J., Michiels, L., Derdelinckx, G., Voordeckers, K., and Verstrepen, K. (2015). Purification and characterization of a novel Brettanomyces anomalus beta-glucosidase enzyme suitable for food bioflavoring. unpublished. White, C. and Zainasheff, J. (2010). Yeast: the practical guide to beer fermentation. Brewers Publication.

Woolfit, M., Rozpedowska, E., Piškur, J., and Wolfe, K. (2007). genome Survey Sequencing of the Wine Spoilage Yeast Dekkera (Brettanomyces) bruxellensis. Eukaryotic Cell, 6(4):721–733.

Yakobson, C. (2010). Brettanomyces project.