Screening of Brazilian isolated yeasts reveals multiple potential chassis for industrial biotechnology

Screening of Brazilian isolated yeasts reveals multiple potential chassis for industrial biotechnology | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 29 October 2025 V1 Latest version Share on Screening of Brazilian isolated yeasts reveals multiple potential chassis for industrial biotechnology Authors : Giovanna R. Maklouf 0000-0003-2213-1899 , Lara I. S. Sousa , Juliana Galhardo , Cecilia Trivellin 0000-0003-4860-9790 , João Pedro Rodrigues Prado , Tássia Cristina da Silva , Ana C. S. R. de Carvalho , Juliana José , Goncalo Pereira 0000-0003-4140-3482 , Marcelo F. Carazzolle [email protected] , and Fellipe da Silveira Bezerra de Mello 0000-0001-5842-1682 Authors Info & Affiliations https://doi.org/10.22541/au.176169973.32378852/v1 321 views 223 downloads Contents Abstract 1. Introduction 2. Materials and Methods 2.7. Correlation analysis of phenotypic responses 2.8. Performance and robustness estimation 3. Results 3.2.2. Feedstock-derived toxic compounds Discussion Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract The future of sustainable industrial biotechnology relies on microbial hosts that can thrive where conventional strains fail. Although Saccharomyces cerevisiae remains the workhorse of industrial production, its intrinsic metabolic constraints and susceptibility to multiple stresses underscore the need for alternative yeast chassis. Here, we present a comprehensive characterization of 79 yeast isolates from the Brazilian Yeast Collection (BRYC), sourced from diverse ecosystems across six Brazilian biomes. Strains were evaluated for growth under industrially relevant conditions, including alternative carbon sources, inhibitory compounds from biomass pretreatment, elevated temperatures, and high ethanol concentrations. Phenotyping was carried out in both solid and liquid media, allowing comparison of trait analyses across cultivation conditions, and revealing how growth environment influences phenotypic expression. Performance was further assessed in raw and hydrolyzed sugarcane and agave biomass. This high-throughput approach identified multiple non- Saccharomyces yeasts with superior traits compared to S. cerevisiae controls. Pichia kudriavzevii (BRYC98) exhibited exceptional multi-stress tolerance, including thermotolerance and resistance to aldehydes, ethanol, and acetic acid. Candida flosculorum (BRYC21), a poorly characterized native yeast species, exhibited remarkable versatility in substrate utilization. Debaryomyces hansenii strains showed the best growth in saponin-rich agave leaf juice - the most challenging substrate tested-, while Wickerhamomyces anomalus strains consistently outperformed controls in sugarcane molasses and hydrolysate. Overall, around 68% of the non-conventional strains exhibited higher robustness across the 15 tested conditions for both maximum biomass and specific growth rate compared to S. cerevisiae . These findings expand the portfolio of stress-resilient, metabolically versatile yeasts and provide a foundation for their development as alternative industrial chassis. Screening of Brazilian isolated yeasts reveals multiple potential chassis for industrial biotechnology Giovanna R. Maklouf a , Lara I. S. Sousa a , Juliana Pimentel Galhardo a , Cecilia Trivellin b,c , João Pedro Rodrigues Prado a , Tássia Cristina da Silva a , Ana C. S. R. de Carvalho d , Juliana José e , Gonçalo A. G. Pereira a , Marcelo F. Carazzolle a* , Fellipe S. B. de Mello a a Laboratory of Genomics and Bioenergy (LGE), Institute of Biology, University of Campinas (Unicamp), Campinas, SP, Brazil b Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, MA, US c Department of Chemistry and Molecular Biology, Gothenburg University, Gothenburg, Sweden d Laboratory of Bees, Biotechnology and Sustainability Studies, University of São Paulo (USP), São Paulo, SP, Brazil e Laboratory of Genetics and Molecular Cardiology (LGCM), InCor, HC-FMUSP, University of São Paulo (USP), São Paulo, SP, Brazil. *corresponding author Keywords: industrial microbiology, biodiversity, non-saccharomyces, robustness Take Away • 79 NCYs were isolated in different biomes and ecosystems in Brazil. • Solid vs liquid screening reveals distinct trait responses to cultivation status. • P. kudriavzevii strains excels in industrial media and tolerates multiple inhibitors. • Undercharacterized yeasts ( C. flosculorum, S. cellae ) display metabolism flexibility. • NCYs show superior stress tolerance and robustness over S. cerevisiae controls. Abstract The future of sustainable industrial biotechnology relies on microbial hosts that can thrive where conventional strains fail. Although Saccharomyces cerevisiae remains the workhorse of industrial production, its intrinsic metabolic constraints and susceptibility to multiple stresses underscore the need for alternative yeast chassis. Here, we present a comprehensive characterization of 79 yeast isolates from the Brazilian Yeast Collection (BRYC), sourced from diverse ecosystems across six Brazilian biomes. Strains were evaluated for growth under industrially relevant conditions, including alternative carbon sources, inhibitory compounds from biomass pretreatment, elevated temperatures, and high ethanol concentrations. Phenotyping was carried out in both solid and liquid media, allowing comparison of trait analyses across cultivation conditions, and revealing how growth environment influences phenotypic expression. Performance was further assessed in raw and hydrolyzed sugarcane and agave biomass. This high-throughput approach identified multiple non- Saccharomyces yeasts with superior traits compared to S. cerevisiae controls. Pichia kudriavzevii (BRYC98) exhibited exceptional multi-stress tolerance, including thermotolerance and resistance to aldehydes, ethanol, and acetic acid. Candida flosculorum (BRYC21), a poorly characterized native yeast species, exhibited remarkable versatility in substrate utilization. Debaryomyces hansenii strains showed the best growth in saponin-rich agave leaf juice - the most challenging substrate tested-, while Wickerhamomyces anomalus strains consistently outperformed controls in sugarcane molasses and hydrolysate. Overall, around 68% of the non-conventional strains exhibited higher robustness across the 15 tested conditions for both maximum biomass and specific growth rate compared to S. cerevisiae . These findings expand the portfolio of stress-resilient, metabolically versatile yeasts and provide a foundation for their development as alternative industrial chassis. 1. Introduction Escalating resource consumption and carbon cycle disruption are hallmarks of our petroleum-dependent industrial age and demand a paradigm shift toward environmentally responsible production strategies. Microbial platforms, particularly yeasts, hold immense promise for bio-based synthesis of valuable compounds from renewable sources, especially biomass and agro-industrial residues. In the context of biorefineries, a critical factor for maintaining efficient and commercially viable fermentation processes lies in the pleiotropic capability of the yeast to co-utilize and catabolize a diverse range of sugars while enduring fermentation stress. S. cerevisiae has been traditionally considered the workhorse in industrial fermentation, due to its ethanol-tolerant and Crabtree-positive traits (Piskur et al., 2006). Nevertheless, its preference for hexose sugars, limited capacity to accumulate intracellular lipids, and low tolerance to inhibitors present in lignocellulosic hydrolysates exemplify the constraints that limit S. cerevisiae ’s applicability in other relevant bioconversion processes. Therefore, identifying non-conventional yeasts (NCYs) and assessing their potential as alternative chassis for industrial biotechnology are essential for the expansion of sustainable production (Geijer et al., 2022). Different species are selected for one or more intrinsic abilities advantageous to a given process, such as an endogenous substrate metabolization pathway, flexible substrate metabolization, an effective secretion system, advanced post-translational modification machinery, stress tolerance or native molecule production. Several NCYs, including Yarrowia lipolytica (Park & Ledesma-Amaro, 2023), Pichia pastoris (Ergün et al., 2022; Love et al., 2018), Kluyveromyces lactis (Spohner et al., 2016), have already demonstrated their potential in industry (Thorwall et al., 2020). Strain robustness, defined as the ability to withstand diverse perturbations while maintaining stable growth performance, is a critical determinant of successful industrial applications, as variations in substrate composition and the requirement for stable production directly affect economic viability (Olsson et al., 2022). However, the complex and often polygenic nature of these robust phenotypes in non-conventional organisms, coupled with the incomplete understanding of the underlying molecular mechanisms, often hinders their direct transfer into common hosts like E. coli and S. cerevisiae (Thorwall et al., 2020). Despite this, while genetic manipulation of NCYs can be more challenging, these strains may require fewer modifications to achieve effective large-scale production (Thorwall et al., 2020). This robustness advantage is particularly necessary because fermentation processes subject yeasts to cellular stresses that compromise performance through diminished viability, metabolic overflow to byproducts, excessive protease secretion, and formation of undesired compounds that negatively affect product quality (Walker & Basso, 2020). These stresses are relevant across all biomass-based industries and arise from three primary sources: production processes, feedstocks, and metabolism (Mohedano et al., 2022). Production process-related stresses contribute to cellular compromise through either elevated temperatures or low-pH, often stemming from inadequate cooling systems and bacterial contamination, respectively. Employing naturally thermotolerant or acid-resistant microbes would enable a more sustainable simultaneous saccharification and fermentation operations and would reduce costs related to temperature and pH control. Biomass hydrolysis for substrate availability also releases inhibitory compounds for yeast metabolism, underscoring the value of utilizing tolerant strains. Specifically, tolerance toward weak acids, such as acetic acid, and aldehydes, like furfural and 5-hydroxymethylfurfural (HMF), is critical, as these are commonly generated from sugar degradation. These compounds, along with other inhibitory molecules, can significantly impair microbial intracellular processes, leading to reduced biosynthesis yields (Almeida et al., 2011). Finally, beyond external stresses from processes and feedstocks, metabolic byproducts produced during bioprocessing, including ethanol, can further impact cellular physiology. Bearing this in mind, we sought to investigate novel NCYs isolated from Brazilian ecosystems. Brazil, with its vast territory encompassing six distinct biomes characterized by diverse landscapes, geomorphology, soils, fauna, and flora, presents substantial opportunities for discovering new microbial resources (Ellwanger et al., 2022). To assess their industrial potential, we evaluated Brazilian yeast isolates for key fermentation characteristics: thermotolerance, resistance to organic acids, aldehydes, and ethanol, as well as their ability to thrive in low pH environments and utilize various carbon sources across both solid and liquid media. We also tested these strains in raw industrial media using sugarcane and agave lignocellulosic hydrolysates, sugarcane molasses, and agave juice. Our findings identified several promising strains — some previously recognized for their robustness and existing genetic modification tools, and others that remain largely unexplored, thereby opening new avenues for industrial application. These results demonstrate the potential of such yeasts to enhance fermentation processes and contribute to biotechnological advancements. 2. Materials and Methods A total of 79 yeast strains were analyzed in this study. A complete description of the collection, including strain identification numbers, taxonomic assignments (where available), and the geographical or environmental origin of each isolate, is provided in Supplementary Table S1 . To establish a benchmark, six S. cerevisiae control strains were included in the study: Control C1 (ACY503), known for its tolerance to low pH (Coradini et al., 2021); Control C2 (BRB), a wine fermentation isolate (Peter et al., 2018) resistant to acetic acid (in-house data); Control C3 (BY4742), a S288c derivative yeast, a widely used laboratory strain (Brachmann et al., 1998); Control C4 (FMY097), an industrial segregant with aldehyde tolerance (de Mello et al., 2019); Control C5 (LVY34.4), a modified industrial strain for xylose fermentation (dos Santos et al., 2016); and Control C6 (PE-2), an industrial sugarcane fermentation isolate renowned for its hexose fermentation capabilities (Basso et al., 2008). All strains were routinely maintained on YPD medium composed of 10 g L⁻¹ yeast extract (Kasvi), 20 g L⁻¹ peptone (Neogen), and 20 g L⁻¹ glucose (Sigma-Aldrich). For long-term preservation, isolates were stored as glycerol stocks (70% v/v) at –80 °C. Genomic DNA was obtained using the Yeast DNA Extraction Kit (Thermo Fisher Scientific). The ITS1-4 region was accessed by Sanger sequencing using the primers ITS1 (5’ TCCGTAGGTGAACCTGCGG 3’) and ITS4 (5’ TCCTCCGCTTATTGATATGC 3’). Sequencing was performed by the BPI Tecnologia Facility (Botucatu, SP, Brazil). Initial identification was performed against the UNITE database version 10.0 (https://unite.ut.ee/ accessed on 07/03/2024), using the database’s best hit. For further corroboration, an in silico phylogenetic inference was performed. The ITS sequences were enriched using data from the Y1000+ database, which comprises the genome sequence of 1,154 yeast strains (Hittinger et al., 2025; Opulente et al., 2024). All BRYC ITS1 sequences were initially blasted against the Y1000 dataset using an e-value of e-20 and the parameter -max_target_seqs 5. Redundancies were discarded, and a final dataset was obtained comprising 203 sequences ( Supporting Information S2 ). Nucleotide alignment was obtained with MAFFT v7.2.271 (Yamada et al., 2016) using –leavegappyregion and 1,000 iterations. IQ-tree v.2.1.2 (Minh et al., 2020) was used for the amino acid substitution best-fit model test (TVM+F+R6 chosen according to BIC), and for phylogenetic inference, which was obtained with 1,000 bootstraps for branch support ( Supporting Information S3 ). Schizosaccharomyces pombe was used as the outgroup, and species identity was corroborated based on the branch pattern. Isolates without phylogenetic corroboration were indicated as incertae sedis . Yeast strain phenotyping was performed in media based on the standard yeast peptone (YP) formulation (1% yeast extract, 2% peptone). For solid media testing, 2% agar was added to each medium. For carbon source utilization assays, YP medium was supplemented with 2% (w/v) of the following substrates: glucose, xylose, maltose, galactose, glycerol (all sourced from Sigma-Aldrich), or inulin (food-grade, sourced from Ingredientes Online ). For stress tolerance assays, YPD (YP supplemented with 2% glucose) was used as the basal medium and further conditioned with one of the following stressors: 12 g/L acetic acid (Sigma-Aldrich) at pH 4.7; pH 2.1 adjusted with H₂SO₄ (LabSynth); 12 and 14% ethanol (Nuclear); 40 mM (5.04 g/L) 5-hydroxymethylfurfural (HMF) (Sigma-Aldrich); 40 mM (3.84 g/L) furfural (Sigma-Aldrich). The high-temperature tolerance assays (40°C and 42°C) were conducted in normal YPD medium. Carbon source media were sterilized by autoclaving at 121°C for 20 min. Media containing inulin, chemical inhibitors or ethanol were sterilized by filtration (0.22 µm), to prevent their hydrolysis, degradation, or evaporation during autoclaving. All media were freshly prepared prior to each experiment, and pH was adjusted after the addition of stressors when necessary. YPD without additives served as the control condition. 2.4. Raw biomass feedstock For the sugarcane-based industrial media, substrates were supplied by partner companies. Type B sugarcane molasses (SCM), obtained from the Maharashtra mill (India) and supplied by Lesaffre, was diluted with water at a ratio of 1:4. Sugarcane lignocellulosic hydrolysate (SCH), from pre-treated sugarcane straw (steam explosion) provided by GranBio S.A., was prepared as described previously (dos Santos et al., 2024), and used without further dilution. To sterilize both SCM and SCH, solid residues and microorganisms were removed through centrifugation and subsequent filtration with 0.22 μm filters. For the agave-based media, all processing was performed in-house. Two fractions of Agave hybrid 11648 (Raya et al., 2023) leaves, collected in the city of Valente (Latitude 11° 22’ 57.637” S and Longitude: 39° 26’ 39.667” W, in Bahia, Brazil), were employed as substrates for hydrolysis experiments: defibration waste (mucilage, also known as pulp) and intact leaves. The mucilage fraction was obtained directly from field decortication of agave leaves, following the mechanical defibration process. Intact leaves were sectioned into fragments of approximately 20 × 20 cm, mechanically comminuted in a knife mill, and subsequently pressed to extract the aqueous fraction – to generate the medium called agave leaf juice (ALJ). The mucilage and leaves were subjected to oven-drying at 65°C for five days to ensure moisture removal. The dried materials were then milled using a Willye knife mill (model TE-680; Tecnal, Piracicaba, Brazil) and sieved to obtain particles with a uniform granulometry of 20 mesh. Direct enzymatic hydrolysis was performed with suspensions containing 25% (w/v) of leaf mucilage and leaf biomass from Agave hybrid 11648, which were prepared in 500 mL Erlenmeyer flasks (da Silva et al., 2025). Cellic CTec2 cellulase cocktail (10 FPU/g biomass, enzyme blend, from Sigma-Aldrich, 310 FPU/mL) was added, and distilled water was used to adjust the final volume of the reaction. The mixtures were incubated in a rotary shaker at 50 °C for 72 h, under constant agitation at 150 rpm. At the end of the hydrolysis, the solid fractions were removed by filtration. These resulting media are called agave mucilage hydrolysate (AMH) and agave leaf lignocellulosic hydrolysate (ALH). The sugar composition and associated inhibitory compounds of agave- and sugarcane-derived feedstocks were determined by high-performance liquid chromatography and are detailed in Table 1 . † - ND - Level (concentration) of the compound was not detected; § NA: not applicable; ‡ (—): Not reported The cryopreserved yeast collection was pre-cultured in liquid YPD medium at 30°C overnight without agitation. This procedure was repeated once more to standardize the inoculum prior to phenotyping. For the assays, 30 µL of the pre-culture was transferred into 120 µL of the respective conditioned medium in sterile 96-well flat-bottom microplates (SpectraPlate®, PerkinElmer). Plates were sealed with translucent, gas-impermeable adhesive films (MicroAmp™, Applied Biosystems) to minimize evaporation and contamination. Each condition was tested in triplicate (technical replicates). Growth measurement was performed using an automated high-throughput platform. Plates were incubated at 30°C in a Cytomat (Thermo) automated incubator. A robotic handler MICROLAB STARlet (Hamilton) periodically transferred plates to a SpectraMax Plus 384 (Molecular Devices) spectrophotometer to measure absorbance at 600 nm every 180 minutes for growth quantification. The entire sequence, linking the incubator, spectrophotometer, and robotic handler, was controlled by a Python script integrated using PyHamilton. Raw optical density (OD) measurements from yeast growth experiments were exported from Excel files containing multiple sheets, each corresponding to one of the three replicates. Custom R scripts (2023.06.1+524), using the packages tidyverse (v2.0.0) and ggplot2 (v3.5.1), were used to: (i) import and combine replicate sheets, aligning sample identifiers and time points; (ii) remove blanks; (iii) normalize sample names to match a reference strain list; and (iv) calculate summary statistics, including mean OD and standard deviation at each time point. Top-performing strains were identified based on maximum mean OD, and growth curves and dot plots were generated to visualize these results. Pre-inoculum preparation was carried out following the procedures described in Section 2.6. The strains were then inoculated onto agar plates containing the respective carbon sources and stressors using a 96-pin replicator (Boekel Scientific). Photographic records of the plates were obtained after 3 days of growth to capture the final endpoint phenotype. Colony size was subsequently quantified using the ImageJ software in conjunction with the microplate analysis plugin developed by Jay Unruh (Stowers Institute for Medical Research, Kansas City). The resulting colony size values were corrected by subtracting the blank values (negative controls) and were then used to calculate the z-score for each strain per condition/plate, providing the standardized metric for growth performance in solid media. 2.7. Correlation analysis of phenotypic responses All correlation analyses were performed using the Spearman rank correlation coefficient (r), which assesses the strength and direction of the monotonic relationship between the ranked growth responses of the yeast strains. The data used for these analyses were derived from the two distinct screening platforms: liquid culture (metric: maximum OD value, Max OD) and solid culture (metric: colony size). Both raw metrics were normalized using the z-score calculation across all strains for each condition individually to ensure comparability. To assess the degree to which growth profiles under different substrates or stressors were associated within the same cultivation format, intra-medium correlation matrices were generated. The Spearman coefficient was calculated between the z-score vectors of all pairwise conditions (e.g., Xyl vs. T_40°C). This process was performed separately for the Max OD z-scores (liquid assays) and the colony size z-scores (solid assays). The resulting matrices were organized by hierarchical clustering to group conditions with similar response patterns. Statistical significance was determined using a p-value threshold, with alpha = 0.05. Two distinct methods were employed to compare the consistency of phenotypic responses between the liquid and solid media. To directly assess the consistency of a strain’s overall performance across the two media types, a strain-specific correlation was computed. For each individual yeast strain, the vector of its Max OD z-scores (liquid medium) was aligned and paired with the corresponding vector of its colony size z-scores (solid medium) across all common conditions. The Spearman coefficient was then calculated between these paired vectors, quantifying the stability of the strain’s phenotypic ranking regardless of the culture format. A condition-specific correlation analysis was performed to determine which individual conditions resulted in concordant strain performance rankings across both the liquid and solid media assays. For each condition, the Spearman coefficient was calculated between the vector of all strains’ liquid medium z-scores (Max OD) and the vector of the same strains’ solid medium z-scores (colony size). The resulting Spearman coefficient quantified the concordance of strain performance for that specific condition. Data processing, normalization, and statistical analyses were performed using the R statistical environment and associated packages, including tidyverse (v2.0.0) and corrplot (v0.92). 2.8. Performance and robustness estimation Performance was evaluated based on maximum biomass (OD) and maximum specific growth rate (1/h). Maximum biomass was defined as the highest OD value reached during cultivation. The maximum specific growth rate (μmax) was obtained as the peak value of the first derivative of smooth spline curves fitted to OD time-course data for each strain, replicate, and condition, using the all_splines() function with a smoothing parameter of 0.7 (Sprouffske & Wagner, 2016). Robustness was calculated following Eq. 1 (Olsson et al., 2022), which normalizes the index of dispersion of the data across conditions by the mean value of the data across all strains and conditions. A robustness value of zero represents the theoretical maximum.\(R=\ -\left(\frac{\sigma^{2}}{\overline{x}}\frac{1}{m}\right)\)Eq. 1 The percentage of strains with robustness greater than the control strains (average of C1–C7) was calculated as the number of such strains multiplied by 100 and divided by the total number of strains. Robustness was also assessed within grouped conditions: the industrial group (sugarcane molasses, sugarcane lignocellulosic hydrolysate, agave leaf juice, agave mucilage hydrolysate, and agave leaf lignocellulosic hydrolysate), the stressors group (HMF, furfural, acetic acid, low pH, high temperature, and ethanol), and the sugars group (glucose, maltose, glycerol, xylose, inulin, and galactose). 3. Results Sample and morphological characterization of the isolated yeast collection Yeasts collected from diverse locations across Brazil were screened for fermentative capacity in YPD medium, yielding 79 isolates that comprise the Brazilian Yeast Collection (BRYC) ( Supporting Information S1 ). Sample sources included fruits (mangaba [ Hancornia speciosa ], pineapple [ Ananas comosus ]), plants and plant products ( Salicornia spp ., cumaru seeds [ Dipteryx odorata ], Araucaria angustifolia ), bee-associated materials (honey, propolis) and bee species (jataí stingless bee, Euglossa, Apis bees), and fermented foods (kefir, beer). Geographic sampling extended from Brazil’s southern coast to southeastern urban areas. Colony morphology on standard YPD medium is illustrated in Figure S1 . Each isolate was taxonomically classified to the species or genus level, with detailed geographic provenance documented in the collection. Taxonomic identification was achieved through analysis of the internal transcribed spacer (ITS) region and corroborated by phylogenetic analysis ( Figure 1 ). In instances where phylogenetic classification was inconclusive, the designation ’ incertae sedis ’ was employed to indicate taxonomic ambiguity. Due to unsuccessful amplification or sequencing of the ITS region, some species could not be identified. These species are provisionally labeled with the collection code ’BRYC’ followed by a unique identification number. A notable trend emerged with the overrepresentation of Debaryomyces in bee-associated samples, suggesting a potential ecological adaptation or preference for this genus. Metschnikowia and Starmerella genera were predominantly found in floral samples, while S. cerevisiae was the dominant species in fermented products, aligning with its established role in bakery and beverage fermentation. Similarly, Meyerozyma caribbica and Wickerhamomyces anomalus demonstrated a broad ecological distribution. Overall, Debaryomyces , Saccharomyces , Candida , and Wickerhamomyces were the dominant genera in our collection. While we identified less common species like Candida flosculorum, Starmerella cellae , Starmerella kuoi and Metschnikowia matae , our collection also included well-known industrially valuable yeasts such as Debaryomyces hansenii and Wickerhamomyces anomalus (Arous et al., 2017; Navarrete et al., 2022) . Figure 1. ITS-based phylogeny of BRYC isolates. The phylogenetic inference was reconstructed using the Maximum Likelihood method in IQ-Tree with 1,000 bootstrap replicates to determine branch support. Isolate characteristics are indicated by color and icons: the collection site corresponds to the color code, and the ecological origin is represented by the icons. Branch length is not shown in this representation. 3.2. High-throughput phenotyping reveals multiple stress-tolerant strains To characterize the isolates, a high-throughput screening assay in solid and liquid media under different conditions was performed. Given the inconsistency of phenotyping methods in the literature, the use of different medium conditions was performed to also evaluate the correlation between cultivation state and trait response. The growth of yeasts present in BRYC was examined under four categories of stressors: 1) alternative sugar sources - glucose, xylose, maltose, galactose, glycerol, and inulin, each at a concentration of 2%; 2) feedstock-derived toxic compounds - 40 mM of HMF (5.04 g/L), 40 mM of furfural (3.84 g/L) and acetic acid at 12 g/L, with pH maintained at 4.7; 3) intrinsic bioprocess stresses - acidic pH (pH equal to 2.1, achieved with H 2 SO 4 ) and temperatures of 40ºC and 42ºC; and 4) high ethanol concentrations of 12% and 14%. YP medium was used to support growth for a broader range of strains in our collection, as several strains exhibited growth limitations in YNB (results not shown). For solid media screening, results are presented as z-scores of relative colony size values ( Figure 2 ), while liquid media results are displayed as growth curves of OD data ( Figures 3 and 4 ) and discussed in the following sections. Figure 2. Performance of BRYC strains on various solid phenotyping conditions. Dot plots show the relative colony size (z-score) for all BRYC strains across 13 solid media conditions after 24 hours. The top 10 growing strains are highlighted by colored dots, while the remaining strains are black. The dashed red line marks the reference threshold, representing the highest growth z-score achieved by the S. cerevisiae control strain in each condition. On solid media ( Figure 2a-f ), Pichia kudriavzevii (BRYC98) excelled on glucose, maltose, glycerol, and inulin, while Candida flosculorum (BRYC21) demonstrated superior growth on xylose and galactose, with notable performance on glycerol and inulin as well. In liquid media ( Figure 3 ), Meyerozyma caribbica (BRYC74) exhibited optimal growth on xylose, galactose, and maltose, whereas C. flosculorum (BRYC21) thrived once again on glycerol and inulin. Interestingly, C. flosculorum achieved a faster growth rate than M. caribbica on maltose despite reaching a similar peak cell density. Despite the expectation that the robust C6 (PE-2) strain would rank among the top performers for glucose utilization, none of the S. cerevisiae control strains, including C6, were among the top ten performers for any of the six sugars tested, except for control C5 (LVY34.4) on xylose, an engineered strain optimized through adaptive evolution for xylose consumption (dos Santos et al., 2016). Nevertheless, M. caribbica (BRYC74), BRYC58 (unclassified strain) and C. flosculorum (BRYC21) consistently outperformed the LVY34.4 S. cerevisiae strain. Figure 3. Performance of BRYC strains on different sugars in liquid phenotyping . Growth curve of the top 10 strains and dot plot of top 20 strains on A) glucose, B) xylose, C) maltose, D) galactose, E) glycerol, and F) inulin. In the dot plot, the values on the x-axis represent the z-score calculated from the maximum OD value of each strain (in triplicate) for each condition. The colored dots denote the Top 10 growing strains, while the gray dots represent the remaining strains. The dashed black line indicates a reference threshold (the z-score of the S. cerevisiae control strain that exhibited the highest growth in each condition). 3.2.2. Feedstock-derived toxic compounds Furfural and 5-hydroxymethylfurfural (HMF) are primary furanic aldehydes generated during biomass pretreatment. Under solid conditions, P. kudriavzevii (BRYC98) demonstrated superior tolerance to both furfural and HMF ( Figure 2i-j ). However, W. anomalus (BRYC03) exhibited slightly higher growth than P. kudriavzevii (BRYC98) in the presence of furfural ( Figure 2i ). Additionally, Pichia fermentans (BRYC08) and C. methanosorbosa (BRYC14) also displayed tolerance to HMF. In liquid culture, Candida flosculorum (BRYC21) demonstrated the highest tolerance to HMF ( Figure 4a ), and P. fermentans (BRYC08), BRYC18 (unclassified strain), and C. flosculorum (BRYC21) exhibited similar levels of tolerance to furfural ( Figure 4b ) over the five-day experimental period. Our control C4 strain (FMY097, a highly aldehyde tolerant strain) for these two conditions showed strong growth only in solid furfural medium, ranking last in the top 10. We also investigated the effects of weak acids, particularly acetic acid, which is the most prevalent in lignocellulosic hydrolysates. In our assessment, P. kudriavzev ii demonstrated superior performance on solid media, followed by P. membranifaciens ( Figure 2g). While the S. cerevisiae control C2 strain (BRB) ranked among the top three performers in solid media, its performance did not translate to liquid media. In liquid culture, Wickerhamomyces anomalus (BRYC30) emerged as the most robust strain over a 5-day period ( Figure 4c ). However, P. kudriavzevii reached peak growth in just 17 hours, whereas W. anomalus took 59 hours to achieve a similar peak. After 17 hours, P. kudriavzevii exhibited a decline in growth, possibly due to the rapid consumption of substrates in the medium. Thus, P. kudriavzevii stood out for presenting high tolerance to acetic acid in both solid and liquid media. Figure 4. Performance of BRYC strains on different stressors in liquid phenotyping . Growth curve of the top 10 strains and dot plot of top 20 strains on A) HMF 40mM, B) Furfural 40mM, C) Acetic acid 12 g/L, D) Acidic pH (pH =2), E) Growing temperature of 40ºC, F) Growing temperature of 42ºC, G) Ethanol 12% and H) Ethanol 14%. In the dot plot, the values on the x-axis represent the z-score calculated from the maximum OD value of each strain (in triplicate) for each condition. The colored dots denote the Top 10 growing strains, while the black dots represent the remaining strains. The dashed red line indicates a reference threshold (the z-score of the S. cerevisiae control strain that exhibited the highest growth in each condition). 3.2.3. Intrinsic bioprocess stresses: temperature and pH Low pH environments may occur naturally due to acid accumulation or be deliberately maintained to minimize bacterial contamination in bioprocesses. In our tests, W. anomalu s (BRYC03), P. kudriavzevii , and P. fermentans (BRYC08) exhibited superior performance on solid media under low pH conditions ( Figure 2h ). In contrast, BRYC58 (unclassified strain) and M. caribbica (BRYC74) demonstrated the most favorable growth curves in liquid media at pH = 2.1 ( Figure 4d ). Among the control strains, none, including control C1 (ACY503) with its reported low pH tolerance, ranked among the top 20 performers in any growth assay. To assess strain thermotolerance, a critical factor when considering the exothermal nature of some biochemical reactions or the need for biomass hydrolysis downstream of the fermentation, we conducted growth experiments at 40ºC and 42ºC. The temperatures selected aimed to enable comparison with the tolerance levels reported for elite S. cerevisiae strains. P. kudriavzevii (BRYC98) demonstrated exceptional thermotolerance, consistently outperforming all other strains across both solid ( Figure 2l-m ) and liquid ( Figure 4e-f ) media. In contrast, other strains exhibited varying performance. On solid media at 40°C, only Starmerella cellae (BRYC44) demonstrated growth comparable to that of the control strains ( Figure 2l ). However, under liquid conditions at 40°C ( Figure 4e ), a broader range of strains, including S. cellae (BRYC44), Debaryomyces hansenii (BRYC72), Zygosaccharomyces rouxii (BRYC09), and two M. caribbica isolates (BRYC74 and BRYC73), exhibited growth comparable to or exceeding that of the control S. cerevisiae strains (C4 and C5, derived from bioethanol industry strains). While these strains demonstrated robust performance across both solid and liquid media, they still lagged behind the thermotolerance of P. kudriavzevii (BRYC98). 3.2.4. Ethanol tolerance Ethanol is a prime example of yeast metabolism, which could be either the final fermentation product or a byproduct, especially when using biomass as a raw material. Therefore, assessing yeast tolerance to this molecule is also imperative. Strain growth was assessed in media containing 12% and 14% ethanol ( Figure 4g-h ). Ethanol tolerance was tested exclusively in liquid medium. The results corroborated the well-known ethanol tolerance of S. cerevisiae , with several control strains ranking among the top 10 for best growth. Four control strains were top performers at 12% ethanol concentration, and two maintained this ranking at 14% ethanol. Notably, BRYC01, a wild S. cerevisiae isolated from the Salicornia plant within our collection, exhibited optimal growth at both 12% and 14% ethanol. In addition to BRYC01, Meyerozyma caribbica (BRYC32), Starmerella apicola (BRYC33), and Debaryomyces hansenii (BRYC34) also demonstrated robust growth at both ethanol concentrations. 3.2.5. Industrial scenarios To evaluate strain performance in industrial media, we utilized raw and lignocellulosic substrates derived from sugarcane and agave. While sugarcane is already established as a bioenergy crop, agave emerges as a promising biomass source under climate change scenarios due to its adaptability to semi-arid regions ( Raya et al., 2023). Both biomasses could be used as feedstocks within biorefinery concepts. We employed five industrial media to explore the potential of strains growing in these media: sugarcane type B molasses (SCM), sugarcane lignocellulosic hydrolysate (SCH), agave leaf juice (ALJ), agave mucilage hydrolysate (AMH) and agave leaf lignocellulosic hydrolysate (ALH) ( Figure 5) . Agave mucilage was herein employed as this residue is commonly found in agave fiber processing (da Silva et al., 2025; Ferraz-Almeida et al., 2025). Four strains - P. fermentans (BRYC08), S. cerevisiae (BRYC41), BRYC58 (unclassified strain), and P. kudriavzevii (BRYC98) - demonstrated robust growth in both agave hydrolysates (ALH and AMH) ( Figure 5a-b ). Within ALH specifically, D. hansenii (BRYC19), W. anomalus (BRYC30), and notably P. kudriavzevii (BRYC98) exhibited superior performance, with BRYC98 achieving peak OD within just 20 hours ( Figure 5a ). Both BRYC30 and BRYC98 also performed well with the individual stressors relevant to this industrial medium, such as acetic acid and HMF ( Table 1 ). In AMH, a high-xylose medium, M. guilliermondii (BRYC56) was the top performer and also ranked in the top 20 for xylose utilization, followed by P. fermentans (BRYC08) ( Figure 5b ). The minimally processed ALJ presented a more challenging environment, possibly due to the presence of saponins, which can severely compromise yeast membranes (da Costa et al., 2024), and extremely high acetic acid concentrations ( Table 1 ). D. hansenii (BRYC20) and the control C5 strain (xylose consumer) exhibited the most promising growth, although peak OD values remained below 0.6 ( Figure 5c ). Interestingly, five of the top ten performers in ALJ were identified as D. hansenii strains (BRYC19, BRYC20, BRYC47, BRYC65, BRYC88), with three of these (BRYC20, BRYC65, and BRYC88) also ranking in the top 20 for acetic acid tolerance. The industrial sugarcane media (SCH and SCM) analysis showed that the two W. anomalus strains (BRYC30 and BRYC03) were consistently the top performers across both media ( Figure 5d-e ). While four S. cerevisiae strains were represented among the top ten in both, the W. anomalus isolates exhibited superior growth throughout the experimental period. For SCH specifically, P. kudriavzevii (BRYC98) showed exceptionally rapid growth kinetics, reaching peak OD within 30 hours ( Figure 5d ). Notably, the three top-performing BRYC strains—BRYC03, BRYC30, and BRYC98—all ranked in the top 20 for tolerance to HMF, furfural, and acetic acid, which are the inhibitors present in SCH ( Table 1 ). Figure 5. Performance of BRYC strains on industrial agave and sugarcane media in liquid phenotyping . Growth curve of the top 10 strains and dot plot of top 20 strains on A) Agave Leaf Hydrolysate (ALH), B) Agave Mucilage Hydrolysate (AMH), C) Agave Leaf Juice (ALJ), D) Sugarcane Hydrolysate (SCH) and E) Sugarcane Molasses (SCM). In the dot plot, the values on the x-axis represent the z-score calculated from the maximum OD value of each strain (in triplicate) for each condition. The colored dots denote the Top 10 growing strains, while the black dots represent the remaining strains. The dashed red line indicates a reference threshold (the z-score of the S. cerevisiae control strain that exhibited the highest growth in each condition). Cultivation state reveals inter- and intra-condition trait correlations Discrepancies emerged when comparing results from solid versus liquid media. Particularly, screening on solid media presented limitations in result interpretation — especially for Pichia kudriavzevii (BRYC98) and Pichia membranifaciens (BRYC85). These strains exhibited markedly divergent growth patterns across the two media types, most notably in sugar utilization assays. While liquid media typically provide a more nutrient-rich and homogenous environment that supports rapid growth, these two strains showed the opposite trend. Given their well-documented biofilm-forming capacity (Ruiz-Muñoz et al., 2022), we hypothesize that enhanced biofilm formation on solid media may have led to an overestimation of growth in this condition relative to liquid cultures. To provide a broader comparative perspective, we calculated the correlation between strain growth in solid and liquid media, examining both individual strain performance across all conditions ( Figure S2 ) and the global growth response for each specific condition across all strains ( Figure S3 ). At the individual strain level, the majority of yeasts showed a positive, albeit generally weak, correlation between growth in liquid and solid media, meaning their performance in one medium was only slightly similar to their performance in the other. Specifically, 67% of the strains exhibited a positive correlation above 0.2, with most (32.9%) falling in the 0.20−0.39 range, and only 1.2% (represented by S. cellae BRYC44) showing a very strong correlation (r≥0.8). Conversely, a significant portion of the collection displayed no clear relationship or an inverse relationship between growth environments: 16.5% showed very weak correlations (r<0.2), and another 16.5% exhibited negative correlations. The negatively correlated group repeatedly included strains such as Wickerhamomyces sydowiorum (BRYC93, 95), wild S. cerevisiae (BRYC66, 43, 01), Debaryomyces hansenii (BRYC19, 34, 62), and Meyerozyma guilliermondii (BRYC56, 61). When examining the global correlation for individual conditions across all strains, the relationship between solid and liquid growth was exceptionally weak, indicating a near-complete lack of concordance between the two media types. Only tolerance to HMF yielded a minimal correlation above 0.1. Other conditions—maltose, xylose, 40°C (T40), acetic acid, and pH2.1—showed only trace, barely positive correlations (0.00−0.19). Conversely, growth under conditions like glucose, galactose, glycerol, inulin, furfural, and 42°C (T42) consistently resulted in negative correlations. Figure 6. Spearman correlation matrices of growth responses across all yeast strains under various stress and carbon source conditions (A) Correlations based on growth in liquid media and (B) in solid media. Tested conditions include carbon sources (Xyl: xylose, Gal: galactose, Gly: glycerol, Glu: glucose, Mal: maltose, Inu: inulin), chemical or environmental stressors (HMF: hydroxymethylfurfural, FF: furfural, AA: acetic acid, pH: acidic pH, T_40°C: 40°C, T_42°C: 42°C), and industrial media (SCM: sugarcane molasses; SCH: sugarcane lignocellulosic hydrolysate, ALJ: agave leaf juice, AMH: agave mucilage hydrolysate and ALH: agave leaf lignocellulosic hydrolysate). Statistical significance was assessed with a 95% confidence level (α=0.05 for rejecting the null hypothesis, H0), while correlations with p0.10) were left blank. We also examined correlations in strain performance across conditions within the same cultivation state to assess whether growth profiles under different substrates or stressors were associated — for example, whether strains that grew well under one condition tended to grow well under another, or conversely, whether performance traits were independent. In liquid media ( Figure 6a ), growth across different carbon sources—including glucose, maltose, galactose, inulin, xylose, and glycerol—was strongly correlated (r = 0.6–0.9). Responses to the furan aldehydes furfural and HMF were also highly correlated (r = 0.92). Notably, acidic pH response clustered with specific carbon sources, particularly inulin (r = 0.74) and glycerol (r = 0.73). In contrast, growth under elevated temperatures (40°C and 42°C), while strongly correlated with each other (r = 0.85), showed low correlation with most other conditions, with the exception of a modest correlation between acetic acid and 40°C (r = 0.60). Industrial media (SCM, ALH, AMH, SCH) formed a highly intercorrelated cluster (r = 0.61-0.86) that also associated with ethanol conditions (EtOH12 and EtOH14, which themselves were tightly correlated at r = 0.94), although ALJ behaved independently from this group. Correlation analysis of yeast growth on solid media ( Figure 6b ) revealed both conserved and divergent patterns compared to liquid conditions. Growth across various carbon sources remained strongly correlated (r = 0.59–0.83), similar to the pattern observed in liquid media. In contrast, correlations between tolerance to toxic compounds (furfural, HMF, acetic acid) were generally lower or non-significant on solid media. Non-conventional strains are generally more robust than S. cerevisiae Figure 7. Strain robustness across conditions. Normalized robustness (Eq.1) is plotted on the y axis both for max. specific growth rate (top) and max. biomass (bottom). The x-axis corresponds to each strain in the collection plus the averaged control. The dotted lines on y equal to zero correspond to the maximum theoretical robustness while the colored dotted lines correspond to the robustness of the control strains. The colored rectangles highlight the strains that were significantly discussed in this work. To finalize the comprehensive performance assessment of the strain collection, we rigorously evaluated the critical metric of strain robustness by quantifying two key industrial parameters: maximum biomass and maximum specific growth rate. This analysis was performed across the liquid phenotyping panel. Overall, robustness was higher for max. biomass than for max. specific growth rate across strains ( Figure S4 ). Comparison with the control S. cerevisiae strains revealed that most non-conventional yeasts (~70%) were more robust in terms of max. specific growth rate, whereas only ( Figure 7 ). When robustness was analyzed by condition group (sugars, stressors, and industrial substrates), both parameters followed the same trend: sugars showed the highest robustness, followed by stressors, with industrial substrates displaying the lowest robustness ( Figure S5 ). The ranking of robustness (sugars > stressors > industrial substrates) generally reflects increasing complexity and heterogeneity of the growth environment. The strain with the highest robustness in max. specific growth rate was BRYC13 ( F. magnum ), while BRYC53 ( M. rancensis ) showed the highest robustness in max. biomass. Considering both performance and robustness, BRYC84 (unknown species) represented the best compromise for max. specific growth rate, BRYC98 ( P. kudriavzevii ) for max. biomass, and BRYC41 ( S. cerevisiae incertae sedis ) as the best overall compromise across both parameters. Discussion In natural environments, yeasts encounter both planktonic (liquid) and colony/surface-associated (solid) niches, each selecting for different metabolic properties (Čáp & Palková, 2024). Liquid growth prioritizes rapid cell division, efficient nutrient uptake, and metabolic flux, while surface colonization requires cell adhesion, invasive growth, and the ability to access nutrients through diffusion-limited environments, constraints that become particularly critical under nutrient limitation or stress conditions (Tronnolone et al., 2018). These distinct selective pressures may favor different morphological or physiological strategies, such as enhanced cell adhesion or altered colony architecture. While conventional yeast phenotyping on solid media provides valuable insights (Aguiar-Cervera et al., 2021; Madeira-Jr & Gombert, 2018; Mukherjee et al., 2017; Nwaefuna et al., 2023; Tatay-Núñez et al., 2024), liquid media offer a more physiologically relevant model system for industrial applications, better replicating the conditions encountered in liquid-state bioreactors, which remain the predominant configuration in industrial fermentation. Accordingly, the screening of yeast libraries has increasingly relied on microplate-based liquid assays and plate readers (Maresová & Sychrová, 2007), owing to their widespread availability and capacity for high-throughput phenotyping. Here, we observed a near-complete lack of correlation between solid and liquid media performance when evaluating the performance of all strains in each condition. This divergence may stem from fundamental differences in how cells experience their environment. Consequently, high-throughput strain screening efforts should prioritize liquid-based assays when targeting liquid-phase industrial processes, as solid media rankings may not reliably predict performance. The phenotypic diversity captured through such comprehensive screening demonstrates that while no single organism may suffice to address all bioprocessing challenges across diverse industrial applications, augmenting the pool of industrially relevant fermenters enables the optimization of phenotype-matching tailored to specific demands. This need for microbial diversity is exemplified by the limitations observed in current industrial standards. Despite not evaluating ethanol production, S. cerevisiae ’s hallmark trait, industrial control strains displayed suboptimal growth in most conditions tested. This finding is particularly noteworthy given that these strains collectively represent industrial backgrounds praised for their commercial application (Della-Bianca et al., 2012). For instance, a wild S. cerevisiae isolate, BRYC01, obtained from the Salicornia plant, outperformed these control strains in tolerating high ethanol concentrations. The efficient utilization of mixed sugar substrates remains a significant challenge in industrial fermentation due to its impact on product titers and yields. When testing growth on different sugars, strains Meyerozyma caribbica (BRYC74) and the underexplored Candida flosculorum (BRYC21) demonstrated an outstanding ability to grow on different carbon sources. M. caribbica (BRYC74) had the best growth on xylose and galactose, while C. flosculorum excelled in maltose, glycerol, and inulin. Previous studies have reported M. caribbica as a xylose-utilizing organism, although its capacity for xylose fermentation can vary among strains. Despite this, the potential for xylitol production from this yeast remains promising (Tadioto et al., 2022). M. caribbica has also been described as tolerant to high salinity (Hebbale et al., 2021), and consistent with this stress-tolerant profile, BRYC74 performed well under additional challenging conditions including 40°C, pH 2.1, and furfural stress. Beyond its versatile metabolic performance, C. flosculorum ecological origin is also noteworthy. In this study, the species was isolated from a Heliconius butterfly, highlighting its association with insect hosts. This observation is consistent with the earlier single report that identifies C. flosculorum in Heliconia flower bracts in the Atlantic Forest biome of Brazil (Rosa et al., 2007). These ecological associations suggest a potential symbiotic relationship between the yeast and its floral/insect hosts. Both hosts provide a consistent sugar-rich environment, benefiting the yeast’s growth. C. flosculorum phenotyping is first described in this work, where it demonstrated exceptional stress tolerance, exhibiting robust growth on HMF, furfural, and the ability to metabolize diverse sugar sources, including xylose and galactose. Altogether, these molecules are found in hemicellulosic derivatives, suggesting a good chassis for hydrolysate fermentation. Understanding the mechanisms underpinning the superior sugar utilization of M. caribbica and C. flosculorum would benefit the development of microbial cell factories capable of effectively fermenting complex sugar mixtures derived from plant biomass (Gao et al., 2019; Kim et al., 2012). Another notable strain with limited characterization is Starmerella cellae , previously associated with solitary bees (Pimentel et al., 2005). In the present study, S. cellae was again isolated from a solitary bee, specifically one from the genus Euglossa . To the best of our knowledge, this is the first report of this strain being tolerant to high temperatures up to 40°C. Together with the remarkable traits exhibited by C. flosculorum , S. cellae thermotolerance highlights the significant knowledge gap in diverse yeast species that could be tailored for industrial applications. While trait analysis in isolated conditions is important to underline relevant phenotypes in different yeasts, characterization of the performance in raw feedstocks is foremost when envisioning industrial applications. Biomass-derived substrates are known for their complex composition, and the interplay of multiple inhibitors can differentially affect cells when combined. Here, we employed sugarcane and agave-derived media to evaluate the NCYs growth under such circumstances. Sugarcane is the predominant crop for bioethanol production, valued for its high biomass yield (de Souza et al., 2013). Conversely, agave, traditionally cultivated for beverage and fiber production, has emerged as a promising alternative (de Lourdes Pérez-Zavala et al., 2020). This desert-adapted plant offers potential as a sustainable feedstock for biorefineries (Raya et al., 2022). In this context, NCYs emerge as promising biocatalysts due to their broader metabolic and physiological diversity. Pichia kudriavzevii (BRYC98) consistently ranked among the top 10 performers in all industrial media evaluated, except for ALJ. The strain had already been described as the second most prevalent species in identifications of agave microbiota strains, which may be indicative of this tolerance (Gallegos-Casillas et al., 2024). For instance, in ALH, it reached near-maximum experimental OD in just 20 hours. Similarly, in SCH, it achieved the highest OD within 30 hours, significantly outpacing the second-best strain. Individual stressor analysis consistently identified P. kudriavzevii (BRYC98) as a high-performing strain, demonstrating exceptional tolerance to acetic acid, HMF, 12% ethanol, and elevated temperatures. The rapid growth kinetics exhibited by P. kudriavzevii across diverse stress conditions and substrates underscore its strong potential for industrial applications, as reflected by its extensive current use (Sun et al., 2020; Tran et al., 2023; Xi et al., 2023). Its efficient substrate-to-biomass conversion promises higher productivity and shorter fermentations, though potential trade-offs in yield or by-product formation remain to be assessed. Metabolic flux analysis could reveal the mechanisms behind its rapid growth, guiding targeted engineering to optimize industrial processes. Beyond P. kudriavzevii , other Pichia species, including P. fermentans (BRYC08) and P. membranifaciens (BRYC85), exhibited notable tolerance to various stressors, such as toxic compounds, low pH, and elevated temperatures. These strains also demonstrated robust growth in industrial media, suggesting a general resilience within the Pichia genus. Among the industrial media used, agave leaf juice presented the most significant challenges for microbial growth, as evidenced by the lowest maximum OD values across experiments. The observed growth limitation can be attributed to elevated saponin concentrations characteristic of this feedstock (da Costa et al., 2024). Despite these challenges, Debaryomyces hansenii strains consistently demonstrated superior performance in ALJ, with five out of ten top performers belonging to this species. This overrepresentation suggests a potential adaptation of D. hansenii to the unique stress factors present in ALJ. While the relationship between D. hansenii and agave plants remains relatively unexplored, previous studies have reported the isolation of D. hansenii from naturally fermented Agave fourcroydes (Lappe-Oliveras et al., 2008). Multiple Wickerhamomyces anomalus strains (BRYC03, BRYC30, and BRYC77) demonstrated robust growth across various stress conditions. While sharing a common species identity, these strains exhibited phenotypic variations, with BRYC03 and BRYC30 consistently outperforming BRYC77, particularly in sugarcane-based media (SCH and SCM), acetic acid, and acidic conditions (pH = 2). These observations align with the established industrial relevance of W. anomalus , which has been extensively characterized for its stress tolerance capabilities, including resistance to osmotic pressure and toxic inhibitors commonly encountered in bioprocessing environments (Sehnem et al., 2020; Turner et al., 2022). However, the intraspecies variation observed among our isolates underscores the importance of strain-level characterization, as phenotypic diversity within species can significantly impact industrial performance. In summary, species like Wickerhamomyces anomalus , Debaryomyces hansenii , and Pichia kudriavzevii have been recognized for their industrial potential, but their full capabilities have yet to be fully realized. Additionally, the discovery of robust tolerance in underexplored strains like Candida flosculorum and Starmerella cellae with exceptional phenotypic traits expands the pool of available microbial cell factories. Beyond demonstrating strong individual stress resistance, non-conventional strains from our collection exhibited superior overall robustness compared to S. cerevisiae control strains in terms of specific growth rate. This dual advantage of high individual performance and multi-stress tolerance suggests that these isolates possess physiological adaptations particularly valuable for industrial bioprocessing, where microorganisms must simultaneously cope with complex inhibitor mixtures, fluctuating substrate compositions, and variable environmental conditions. The findings presented herein provide a foundation for selecting optimal yeast strains for specific industrial applications, such as bioethanol production or biorefinery operations. Key strain selection criteria include substrate utilization breadth, product formation capacity, stress tolerance, and growth kinetics. Future research should focus on enhancing strain performance through genetic engineering and exploring novel cultivation parameters to further optimize bioprocess efficiency. Author Contributions G. R. Maklouf: Data Curation, Investigation, Resources, Formal Analysis, Visualization, Writing – Original Draft; L. I. S. Souza: Resources, Investigation; J. P. Galhardo: Formal Analysis; C. Trivelin: Formal Analysis; J. Pedro: Formal Analysis; T. C. da Silva: Resources, Investigation; A. C. S. R. de Carvalho: Investigation, Resources; J. José: Supervision; F. S. B. de Mello: Conceptualization, Supervision, Writing – Review & Editing; M. F. Carazzolle: Conceptualization, Project Administration, Writing – Review & Editing; G. A. G. Pereira: Project Administration. Funding The authors gratefully acknowledge financial support from the São Paulo Research Foundation (FAPESP) for G.R.M. (grant 2023/02363-5) and L.I.S. (grant 2022/07061-4); the Council for the Improvement of Higher Education (CAPES) for G.R.M. (grant 88887.825139/2023-00), L.I.S. (grant 88887.712434/2022-00) and J.P.G. (grant 88887.479699/2020-0); the Novo Nordisk Foundation (DISTINGUISHED INVESTIGATOR 2019, #0055044) and the Swedish Research Council (grant 2024-00345_VR) for C.T. This work was also supported in part by the Center for Computational Engineering and Science (FAPESP/CEPID grant 2013/08293-7) and the National Agency of Petroleum, Natural Gas and Biofuels (Brazil) in association with Shell Brasil Petróleo Ltda, for the Brazilian Agave Development Program (BRAVE) [grant agreements 23018-5/22857-7]. General Acknowledgments The authors gratefully acknowledge Granbio S.A. for the donation of sugarcane hydrolysates and Lesaffre for the donation of molasses used in this study. The authors also thank MSc. Thiago Gaspar Inácio for providing valuable instructions regarding the use of molasses and for sharing the molasses’ sugar composition data. Conflict of Interest Statement The authors declare no conflict of interest. References Aguiar-Cervera, J. E., Delneri, D., & Severn, O. (2021). A high-throughput screening method for the discovery of Saccharomyces and non-Saccharomyces yeasts with potential in the brewing industry. Engineering Biology , 5 (3), 72–80. https://doi.org/10.1049/enb2.12013Almeida, J. R. M., Runquist, D., Sànchez i Nogué, V., Lidén, G., & Gorwa-Grauslund, M. F. (2011). Stress-related challenges in pentose fermentation to ethanol by the yeast Saccharomyces cerevisiae. Biotechnology Journal , 6 (3), 286–299. https://doi.org/10.1002/biot.201000301Arous, F., Atitallah, I. B., Nasri, M., & Mechichi, T. (2017). A sustainable use of low-cost raw substrates for biodiesel production by the oleaginous yeast Wickerhamomyces anomalus. 3 Biotech , 7 (4), 1–10. https://doi.org/10.1007/s13205-017-0903-6Basso, L. C., de Amorim, H. V., de Oliveira, A. J., & Lopes, M. L. (2008). Yeast selection for fuel ethanol production in Brazil. FEMS Yeast Research , 8 (7), 1155–1163. https://doi.org/10.1111/j.1567-1364.2008.00428.xBrachmann, C. B., Davies, A., Cost, G. J., Caputo, E., Li, J., Hieter, P., & Boeke, J. D. (1998). Designer deletion strains derived from Saccharomyces cerevisiae S288C: A useful set of strains and plasmids for PCR-mediated gene disruption and other applications. Yeast , 14 (2), 115–132. https://doi.org/10.1002/(SICI)1097-0061(19980130)14:23.0.CO;2-2Čáp, M., & Palková, Z. (2024). The characteristics of differentiated yeast subpopulations depend on their lifestyle and available nutrients. Scientific Reports , 14 (1), 3681. https://doi.org/10.1038/s41598-024-54300-9Coradini, A. L. V., de Mello F, da S. B., Furlan, M., Maneira, C., Carazzolle, M. F., Pereira, G. A. G., & Teixeira, G. S. (2021). QTL mapping of a Brazilian bioethanol strain links the cell wall protein-encoding gene GAS1 to low pH tolerance in S. cerevisiae. Biotechnology for Biofuels , 14 (1). https://doi.org/10.1186/s13068-021-02079-6da Costa, N. C., dos Santos, M. G. B., de Freitas, F. B. F., da Costa, G. L. V., de Oliveira, G. B., Costa, I. O., de Santana Souza, D. F., dos Santos, E. S., & de Araújo Padilha, C. E. (2024). Effects of sisal saponins on enzymatic hydrolysis and fermentation of lignocellulosic biomass. Bioenergy Research . https://doi.org/10.1007/s12155-024-10798-5da Silva, T. C., Alencar, B. R. A., Soares, A. C. F., Rabelo, S. C., de Sá Barreto Sampaio, E. V., Albuquerque, A. A., & Menezes, R. S. C. (2025). Development of the second-generation ethanol production process from sisal mucilage. Biomass & Bioenergy , 203 (108376), 108376. https://doi.org/10.1016/j.biombioe.2025.108376de Lourdes Pérez-Zavala, M., Hernández-Arzaba, J. C., Bideshi, D. K., & Barboza-Corona, J. E. (2020). Agave: a natural renewable resource with multiple applications. Journal of the Science of Food and Agriculture , 100 (15), 5324–5333. https://doi.org/10.1002/jsfa.10586de Mello, F. da S. B., Coradini, A. L. V., Tizei, P. A. G., Carazzolle, M. F., Pereira, G. A. G., & Teixeira, G. S. (2019). Static microplate fermentation and automated growth analysis approaches identified a highly-aldehyde resistant Saccharomyces cerevisiae strain. Biomass & Bioenergy , 120 , 49–58. https://doi.org/10.1016/j.biombioe.2018.10.019de Souza, A. P., Grandis, A., Leite, D. C. C., & Buckeridge, M. S. (2013). Sugarcane as a Bioenergy Source: History, Performance, and Perspectives for Second-Generation Bioethanol. Bioenergy Research , 7 (1), 24–35. https://doi.org/10.1007/s12155-013-9366-8Della-Bianca, B. E., Basso, T. O., Stambuk, B. U., Basso, L. C., & Gombert, A. K. (2012). What do we know about the yeast strains from the Brazilian fuel ethanol industry? Applied Microbiology and Biotechnology , 97 (3), 979–991. https://doi.org/10.1007/s00253-012-4631-xdos Santos, J. R., Batista, B., de Carvalho, L. M., dos Santos, L. V., Pereira, G., & Carazzolle, M. F. (2024). Assessing the impact of lignocellulose inhibitors on industrial Saccharomyces cerevisiae strains: A comparative study of isomerase and oxidoreductase pathways for xylose metabolism . https://doi.org/10.2139/ssrn.4834432dos Santos, L. V., Carazzolle, M. F., Nagamatsu, S. T., Sampaio, N. M. V., Almeida, L. D., Pirolla, R. A. S., Borelli, G., Corrêa, T. L. R., Argueso, J. L., & Pereira, G. A. G. (2016). Unraveling the genetic basis of xylose consumption in engineered Saccharomyces cerevisiae strains. Scientific Reports , 6 (1), 1–14. https://doi.org/10.1038/srep38676Ellwanger, J. H., Nobre, C. A., & Chies, J. A. B. (2022). Brazilian Biodiversity as a Source of Power and Sustainable Development: A Neglected Opportunity. Sustainability: Science Practice and Policy , 15 (1), 482. https://doi.org/10.3390/su15010482Ergün, B. G., Laçın, K., Çaloğlu, B., & Binay, B. (2022). Second generation Pichia pastoris strain and bioprocess designs. Biotechnology for Biofuels and Bioproducts , 15 (1), 1–19. https://doi.org/10.1186/s13068-022-02234-7Ferraz-Almeida, R., Oliveira, A. R. de, Pinheiro, C. de J., Oliveira, J. L., Almeida, V. F. de, & Arruda, E. M. (2025). Potential of sisal (Agave sisalana) residues for improving sisal plant growth and soil residue stocks in Bahia’s circular agriculture. Agronomy (Basel, Switzerland) , 15 (6), 1426. https://doi.org/10.3390/agronomy15061426Gallegos-Casillas, P., García-Ortega, L. F., Espinosa-Cantú, A., Avelar-Rivas, J. A., Torres-Lagunes, C. G., Cano-Ricardez, A., García-Acero, Á. M., Ruiz-Castro, S., Flores-Barraza, M., Castillo, A., González-Zozaya, F., Delgado-Lemus, A., Molina-Freaner, F., Jacques-Hernández, C., Hernández-López, A., Delaye, L., Aguirre-Dugua, X., Kirchmayr, M. R., Morales, L., … DeLuna, A. (2024). Yeast diversity in open agave fermentations across Mexico. Yeast (Chichester, England) , 41 (1–2), 35–51. https://doi.org/10.1002/yea.3913Gao, M., Ploessl, D., & Shao, Z. (2019). Enhancing the Co-utilization of Biomass-Derived Mixed Sugars by Yeasts. Frontiers in Microbiology , 9 , 409956. https://doi.org/10.3389/fmicb.2018.03264Geijer, C., Ledesma-Amaro, R., & Tomás-Pejó, E. (2022). Unraveling the potential of non-conventional yeasts in biotechnology. FEMS Yeast Research , 22 (1). https://doi.org/10.1093/femsyr/foab071Hebbale, D., Mishra, R. S., & Ramachandra, T. V. (2021). Prioritizing Wild Yeast Strains for Macroalgal Bioethanol Production. Bioenergy Research , 15 (1), 202–217. https://doi.org/10.1007/s12155-021-10283-3Hittinger, C. T., LaBella, A., Opulente, D., Rokas, A., Harrison, M.-C., Wolters, J. F., Liu, C., Kominek, J., Steenwyk, J., Silva, M., Gonçalves, C., Li, Y., Hulfachor, A. B., Zhou, X., Shen, X.-X., Li, Y., & Groenewald, M. (2025). Genomic and ecological factors shaping specialism and generalism across an entire subphylum . Figshare+. https://doi.org/10.25452/FIGSHARE.PLUS.C.6714042Kim, S. R., Ha, S.-J., Wei, N., Oh, E. J., & Jin, Y.-S. (2012). Simultaneous co-fermentation of mixed sugars: a promising strategy for producing cellulosic ethanol. Trends in Biotechnology , 30 (5), 274–282. https://doi.org/10.1016/j.tibtech.2012.01.005Lappe-Oliveras, P., Moreno-Terrazas, R., Arrizón-Gaviño, J., Herrera-Suárez, T., García-Mendoza, A., & Gschaedler-Mathis, A. (2008). Yeasts associated with the production of Mexican alcoholic nondistilled and distilled Agave beverages. FEMS Yeast Research , 8 (7), 1037–1052. https://doi.org/10.1111/j.1567-1364.2008.00430.xLove, K. R., Dalvie, N. C., & Love, J. C. (2018). The yeast stands alone: the future of protein biologic production. Current Opinion in Biotechnology , 53 . https://doi.org/10.1016/j.copbio.2017.12.010Madeira-Jr, J. V., & Gombert, A. K. (2018). Towards high-temperature fuel ethanol production using Kluyveromyces marxianus: On the search for plug-in strains for the Brazilian sugarcane-based biorefinery. Biomass & Bioenergy , 119 , 217–228. https://doi.org/10.1016/j.biombioe.2018.09.010Maresová, L., & Sychrová, H. (2007). Applications of a microplate reader in yeast physiology research. BioTechniques , 43 (5), 667–672. https://doi.org/10.2144/000112620Minh, B. Q., Schmidt, H. A., Chernomor, O., Schrempf, D., Woodhams, M. D., von Haeseler, A., & Lanfear, R. (2020). IQ-TREE 2: New models and efficient methods for phylogenetic inference in the genomic era. Molecular Biology and Evolution , 37 (5), 1530–1534. https://doi.org/10.1093/molbev/msaa015Mohedano, M. T., Konzock, O., & Chen, Y. (2022). Strategies to increase tolerance and robustness of industrial microorganisms. Synthetic and Systems Biotechnology , 7 (1), 533–540. https://doi.org/10.1016/j.synbio.2021.12.009Mukherjee, V., Radecka, D., Aerts, G., Verstrepen, K. J., Lievens, B., & Thevelein, J. M. (2017). Phenotypic landscape of non-conventional yeast species for different stress tolerance traits desirable in bioethanol fermentation. Biotechnology for Biofuels , 10 (1), 1–19. https://doi.org/10.1186/s13068-017-0899-5Navarrete, C., Estrada, M., & Martínez, J. L. (2022). Debaryomyces hansenii: an old acquaintance for a fresh start in the era of the green biotechnology. World Journal of Microbiology & Biotechnology , 38 (6), 1–10. https://doi.org/10.1007/s11274-022-03280-xNwaefuna, A. E., Garcia-Aloy, M., Loeto, D., Ncube, T., Gombert, A. K., Boekhout, T., Alwasel, S., & Zhou, N. (2023). Dung beetle-associated yeasts display multiple stress tolerance: a desirable trait of potential industrial strains. BMC Microbiology , 23 (1), 1–12. https://doi.org/10.1186/s12866-023-03044-zOlsson, L., Rugbjerg, P., Torello Pianale, L., & Trivellin, C. (2022). Robustness: linking strain design to viable bioprocesses. Trends in Biotechnology , 40 (8), 918–931. https://doi.org/10.1016/j.tibtech.2022.01.004Opulente, D. A., LaBella, A. L., Harrison, M.-C., Wolters, J. F., Liu, C., Li, Y., Kominek, J., Steenwyk, J. L., Stoneman, H. R., VanDenAvond, J., Miller, C. R., Langdon, Q. K., Silva, M., Gonçalves, C., Ubbelohde, E. J., Li, Y., Buh, K. V., Jarzyna, M., Haase, M. A. B., … Hittinger, C. T. (2024). Genomic factors shape carbon and nitrogen metabolic niche breadth across Saccharomycotina yeasts. Science (New York, N.Y.) , 384 (6694), eadj4503. https://doi.org/10.1126/science.adj4503Park, Y.-K., & Ledesma-Amaro, R. (2023). What makes Yarrowia lipolytica well suited for industry? Trends in Biotechnology , 41 (2), 242–254. https://doi.org/10.1016/j.tibtech.2022.07.006Peter, J., De Chiara, M., Friedrich, A., Yue, J.-X., Pflieger, D., Bergström, A., Sigwalt, A., Barre, B., Freel, K., Llored, A., Cruaud, C., Labadie, K., Aury, J.-M., Istace, B., Lebrigand, K., Barbry, P., Engelen, S., Lemainque, A., Wincker, P., … Schacherer, J. (2018). Genome evolution across 1,011 Saccharomyces cerevisiae isolates. Nature , 556 (7701), 339–344. https://doi.org/10.1038/s41586-018-0030-5Pimentel, M. R. C., Antonini, Y., Martins, R. P., Lachance, M.-A., & Rosa, C. A. (2005). Candida riodocensis and Candida cellae, two new yeast species from the Starmerella clade associated with solitary bees in the Atlantic rain forest of Brazil. FEMS Yeast Research , 5 (9), 875–879. https://doi.org/10.1016/j.femsyr.2005.03.006Piskur, J., Rozpedowska, E., Polakova, S., Merico, A., & Compagno, C. (2006). How did Saccharomyces evolve to become a good brewer? Trends in Genetics: TIG , 22 (4), 183–186. https://doi.org/10.1016/j.tig.2006.02.002Raya, Fábio Trigo, de Abreu, L. G. F., Marone, M. P., de Araújo Salvador, M., Bressiani, J. A., del Real Laborde, J. I., & Pereira, G. A. G. (2022). New Feedstocks for Bioethanol Production: Energy Cane and Agave. Liquid Biofuels: Bioethanol , 431–455. https://doi.org/10.1007/978-3-031-01241-9_18Raya, F. T., de Carvalho, L. M., José, J., da Cruz, L. P., Almeida, R. L., Delevatti, H. A. de A., Silveira, N. M., da Silva, S. F., Pissolato, M. D., de Oliveira, A. B., dos Reis, W. J. V., de Abreu, L. G. F., Gutiérrez, J., Carazzolle, M. F., Soares, A. C. F., Nieto Sotelo, J., Ribeiro, R. V., & Pereira, G. A. G. (2023). Rescuing the Brazilian Agave breeding program: morphophysiological and molecular characterization of a new germplasm. Frontiers in Chemical Engineering , 5 (1218668). https://doi.org/10.3389/fceng.2023.1218668Rosa, C. A., Pagnocca, F. C., Lachance, M.-A., Ruivo, C. C. C., Medeiros, A. O., Pimentel, M. R. C., Fontenelle, J. C. R., & Martins, R. P. (2007). Candida flosculorum sp. nov. and Candida floris sp. nov., two yeast species associated with tropical flowers. International Journal of Systematic and Evolutionary Microbiology , 57 (Pt 12), 2970–2974. https://doi.org/10.1099/ijs.0.65230-0Ruiz-Muñoz, M., Hernández-Fernández, M., Cordero-Bueso, G., Martínez-Verdugo, S., Pérez, F., & Cantoral, J. M. (2022). Non-Saccharomyces Are Also Forming the Veil of Flor in Sherry Wines. Fermentation , 8 (9), 456. https://doi.org/10.3390/fermentation8090456Sehnem, N. T., Machado, Â. S., Matte, C. R., Morais, M. A. D. E., Jr, & Ayub, M. A. Z. (2020). Second-generation ethanol production by Wickerhamomyces anomalus strain adapted to furfural, 5-hydroxymethylfurfural (HMF), and high osmotic pressure. Anais Da Academia Brasileira de Ciencias , 92 (suppl 2), e20181030. https://doi.org/10.1590/0001-3765202020181030Spohner, S. C., Schaum, V., Quitmann, H., & Czermak, P. (2016). Kluyveromyces lactis: An emerging tool in biotechnology. Journal of Biotechnology , 222 , 104–116. https://doi.org/10.1016/j.jbiotec.2016.02.023Sprouffske, K., & Wagner, A. (2016). Growthcurver: an R package for obtaining interpretable metrics from microbial growth curves. BMC Bioinformatics , 17 (1), 172. https://doi.org/10.1186/s12859-016-1016-7Sun, W., Vila-Santa, A., Liu, N., Prozorov, T., Xie, D., Faria, N. T., Ferreira, F. C., Mira, N. P., & Shao, Z. (2020). Metabolic engineering of an acid-tolerant yeast strain Pichia kudriavzevii for itaconic acid production. Metabolic Engineering Communications , 10 (e00124), e00124. https://doi.org/10.1016/j.mec.2020.e00124Tadioto, V., Milani, L. M., Barrilli, É. T., Baptista, C. W., Bohn, L., Dresch, A., Harakava, R., Fogolari, O., Mibielli, G. M., Bender, J. P., Treichel, H., Stambuk, B. U., Müller, C., & Alves, S. L. (2022). Analysis of glucose and xylose metabolism in new indigenous Meyerozyma caribbica strains isolated from corn residues. World Journal of Microbiology & Biotechnology , 38 (2), 1–14. https://doi.org/10.1007/s11274-021-03221-0Tatay-Núñez, J., Albi-Puig, J., Garrigós, V., Orejas-Suárez, M., Matallana, E., & Aranda, A. (2024). Isolation of local strains of the yeast Metschnikowia for biocontrol and lipid production purposes. World Journal of Microbiology & Biotechnology , 40 (3), 1–14. https://doi.org/10.1007/s11274-024-03918-yThorwall, S., Schwartz, C., Chartron, J. W., & Wheeldon, I. (2020). Stress-tolerant non-conventional microbes enable next-generation chemical biosynthesis. Nature Chemical Biology , 16 (2), 113–121. https://doi.org/10.1038/s41589-019-0452-xTran, V. G., Mishra, S., Bhagwat, S. S., Shafaei, S., Shen, Y., Allen, J. L., Crosly, B. A., Tan, S.-I., Fatma, Z., Rabinowitz, J. D., Guest, J. S., Singh, V., & Zhao, H. (2023). An end-to-end pipeline for succinic acid production at an industrially relevant scale using Issatchenkia orientalis. Nature Communications , 14 (1), 6152. https://doi.org/10.1038/s41467-023-41616-9Tronnolone, H., Tam, A., Szenczi, Z., Green, J. E. F., Balasuriya, S., Tek, E. L., Gardner, J. M., Sundstrom, J. F., Jiranek, V., Oliver, S. G., & Binder, B. J. (2018). Diffusion-limited growth of microbial colonies. Scientific Reports , 8 (1), 5992. https://doi.org/10.1038/s41598-018-23649-zTurner, W., Greetham, D., & Du, C. (2022). The characterisation of Wickerhamomyces anomalus M15, a highly tolerant yeast for bioethanol production using seaweed derived medium. Frontiers in Bioengineering and Biotechnology , 10 , 1028185. https://doi.org/10.3389/fbioe.2022.1028185Walker, G. M., & Basso, T. O. (2020). Mitigating stress in industrial yeasts. Fungal Biology , 124 (5), 387–397. https://doi.org/10.1016/j.funbio.2019.10.010Xi, Y., Xu, H., Zhan, T., Qin, Y., Fan, F., & Zhang, X. (2023). Metabolic engineering of the acid-tolerant yeast Pichia kudriavzevii for efficient L-malic acid production at low pH. Metabolic Engineering , 75 , 170–180. https://doi.org/10.1016/j.ymben.2022.12.007Yamada, K. D., Tomii, K., & Katoh, K. (2016). Application of the MAFFT sequence alignment program to large data-reexamination of the usefulness of chained guide trees. Bioinformatics (Oxford, England) , 32 (21), 3246–3251. https://doi.org/10.1093/bioinformatics/btw412 Figure Legends Figure 1. ITS-based phylogeny of BRYC isolates. The phylogenetic inference was reconstructed using the Maximum Likelihood method in IQ-Tree with 1,000 bootstrap replicates to determine branch support. Isolate characteristics are indicated by color and icons: the collection site corresponds to the color code, and the ecological origin is represented by the icons. Branch length is not shown in this representation. Figure 2. Performance of BRYC strains on various solid phenotyping conditions. Dot plots show the relative colony size (z-score) for all BRYC strains across 13 solid media conditions after 24 hours. The top 10 growing strains are highlighted by colored dots, while the remaining strains are black. The dashed red line marks the reference threshold, representing the highest growth z-score achieved by the S. cerevisiae control strain in each condition. Figure 3. Performance of BRYC strains on different sugars in liquid phenotyping . Growth curve of the top 10 strains and dot plot of top 20 strains on A) glucose, B) xylose, C) maltose, D) galactose, E) glycerol, and F) inulin. In the dot plot, the values on the x-axis represent the z-score calculated from the maximum O.D. value of each strain (in triplicate) for each condition. The colored dots denote the Top 10 growing strains, while the black dots represent the remaining strains. The dashed black line indicates a reference threshold (the z-score of the S. cerevisiae control strain that exhibited the highest growth in each condition). Figure 4. Performance of BRYC strains on different stressors in liquid phenotyping . Growth curve of the top 10 strains and dot plot of top 20 strains on A) HMF 40mM, B) Furfural 40mM, C) Acetic acid 12 g/L, D) Acidic pH (pH =2), E) Growing temperature of 40ºC, F) Growing temperature of 42ºC, G) Ethanol 12% and H) Ethanol 14%. In the dot plot, the values on the x-axis represent the z-score calculated from the maximum OD value of each strain (in triplicate) for each condition. The colored dots denote the Top 10 growing strains, while the black dots represent the remaining strains. The dashed red line indicates a reference threshold (the z-score of the S. cerevisiae control strain that exhibited the highest growth in each condition). Figure 5. Performance of BRYC strains on industrial agave and sugarcane media in liquid phenotyping . Growth curve of the top 10 strains and dot plot of top 20 strains on A) ALH, B) AMH, C) ALJ, D) SCH and E) SCM. In the dot plot, the values on the x-axis represent the z-score calculated from the maximum OD value of each strain (in triplicate) for each condition. The colored dots denote the Top 10 growing strains, while the black dots represent the remaining strains. The dashed red line indicates a reference threshold (the z-score of the S. cerevisiae control strain that exhibited the highest growth in each condition). Figure 6. Spearman correlation matrices of growth responses across all yeast strains under various stress and carbon source conditions (A) Correlations based on growth in liquid media and (B) in solid media. Tested conditions include carbon sources (Xyl: xylose, Gal: galactose, Gly: glycerol, Glu: glucose, Mal: maltose, Inu: inulin), chemical or environmental stressors (HMF: hydroxymethylfurfural, FF: furfural, AA: acetic acid, pH: acidic pH, T_40°C: 40°C, T_42°C: 42°C), and industrial media (SCM: sugarcane molasses; SCH: sugarcane lignocellulosic hydrolysate, ALJ: agave leaf juice, AMH: agave mucilage hydrolysate and ALH: agave leaf lignocellulosic hydrolysate). Statistical significance was assessed with a 95% confidence level (α=0.05 for rejecting the null hypothesis, H0), while correlations with p0.10) were left blank. Figure 7. Strain robustness across conditions. Normalized robustness (Eq.1) is plotted on the y axis both for max. specific growth rate (top) and max. biomass (bottom). The x-axis corresponds to each strain in the collection plus the averaged control. The dotted lines on y equal to zero correspond to the maximum theoretical robustness while the colored dotted lines correspond to the robustness of the control strains. The colored rectangles highlight the strains that were significantly discussed in this work. Supplementary Material File (image1.emf) Download 25.10 MB File (image2.emf) Download 1.98 MB File (image3.emf) Download 2.91 MB File (image4.emf) Download 3.70 MB File (image5.emf) Download 2.08 MB File (image6.emf) Download 1.07 MB File (image7.emf) Download 2.49 MB File (image8.emf) Download 2.69 MB Information & Authors Information Version history V1 Version 1 29 October 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keyword saccharomyces Authors Affiliations Giovanna R. Maklouf 0000-0003-2213-1899 Universidade Estadual de Campinas Instituto de Biologia View all articles by this author Lara I. S. Sousa Universidade Estadual de Campinas Instituto de Biologia View all articles by this author Juliana Galhardo Universidade Estadual de Campinas Instituto de Biologia View all articles by this author Cecilia Trivellin 0000-0003-4860-9790 Harvard University Department of Organismic & Evolutionary Biology View all articles by this author João Pedro Rodrigues Prado Universidade Estadual de Campinas Instituto de Biologia View all articles by this author Tássia Cristina da Silva Universidade Estadual de Campinas Instituto de Biologia View all articles by this author Ana C. S. R. de Carvalho Universidade de Sao Paulo View all articles by this author Juliana José Universidade de Sao Paulo Instituto do Coracao View all articles by this author Goncalo Pereira 0000-0003-4140-3482 Universidade Estadual de Campinas Instituto de Biologia View all articles by this author Marcelo F. Carazzolle [email protected] Universidade Estadual de Campinas Instituto de Biologia View all articles by this author Fellipe da Silveira Bezerra de Mello 0000-0001-5842-1682 Universidade Estadual de Campinas Instituto de Biologia View all articles by this author Metrics & Citations Metrics Article Usage 321 views 223 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Giovanna R. Maklouf, Lara I. S. Sousa, Juliana Galhardo, et al. Screening of Brazilian isolated yeasts reveals multiple potential chassis for industrial biotechnology. Authorea . 29 October 2025. DOI: https://doi.org/10.22541/au.176169973.32378852/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.176169973.32378852/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a0036a83aaf70700',t:'MTc3OTUzMjU0OQ=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();