The interaction between Lactobacillus delbrueckii ssp. bulgaricus M58 and Streptococcus thermophilus S10 can enhanced texture and flavor profile of fermented milk: Insights from metabolomics analysis.

Lactobacillus delbrueckii ssp. bulgaricus M58 (M58) and Streptococcus thermophilus S10 (S10) are 2 dairy starter strains known for their favorable fermentation characteristics. Therefore, this research aimed to study the effects of 1-d low-temperature ripening on the physicochemical properties and metabolomics of fermented milk. Initially, the performance of single (M58 or S10) and dual (M58+S10) strain fermentation was assessed, revealing that the M58+S10 combination resulted in a shortened fermentation time, a stable gel structure, and desirable viscosity, suggesting positive strain interactions. Subsequently, non-targeted metabolomics analyses using LC-MS and GC-MS were performed to comparatively analyze M58+S10 fermented milk samples collected at the end of fermentation and after 1-d low-temperature ripening. The results showed a significant increase in almost all small peptides and dodecanedioic acid in the samples after one day of ripening, while there was a substantial decrease in indole and amino acid metabolites. Moreover, notable increases were observed in high-quality flavor compounds, such as geraniol, delta-nonalactone, 1-hexanol,2-ethyl-, methyl jasmonate, and undecanal. This study provides valuable insights into the fermentation characteristics of the dual bacterial starter consisting of M58 and S10 strains and highlights the specific contribution of the low-temperature ripening step to the overall quality of fermented milk.

Comparison of the Physical and Sensory Properties of Hybrid Citrus Fruit Jaffa® Sweetie in Relation to the Parent Fruits.

In the presented study, an overall Jaffa sweetie evaluation was made to find a correlation between Citrus grandis Osbeck × Citrus paradisi Macf. and its parent fruits' (Citrus grandis Osbeck, Citrus paradisi Macf.) properties. Based on the sensory analysis, it was found that the taste and aroma of the new hybrid fruit are close to pummelo. By the use of chromatographic analysis, the selected monoterpenes present in the fruits were quantified. α-terpineol was typed as the main monoterpene compound in the headspace of sweetie and grapefruit, with the concentrations: 20.96 and 87.9 µg/g, respectively. In turn, γ-terpinene was chosen as the most important monoterpene determining the flavor of sweetie fruit. Based on two-dimensional gas chromatography (GC × GC-TOF-MS) and principal component analysis (PCA) of the data, several volatile compounds were associated with analyzed fruits' aroma. Jaffa Sweetie is the hybrid fruit with sensory properties similar to pummelo with a higher content of monoterpenes, which improves its health benefits compared to the parent fruit. The research presents an instrumental method for assessing the aroma properties of the fruit as a reference method for sensory analysis, commonly used in the industry.

Characterization of the Key Aroma Compounds in Three Truffle Varieties from China by Flavoromics Approach.

The volatile compounds of three different fresh-picked truffle varieties (Tuber sinensis, T1, Tuber sinoalbidum, T2 and Tuber sinoexcavatum, T3) were extracted by headspace solid-phase microextraction (HS-SPME). Separation and identification of volatile components and sulfur compounds were investigated by gas chromatography-olfactometry (GC-O), gas chromatography-mass spectrometry (GC-MS) and gas chromatography with flame photometric detection (GC-FPD). The results showed that 44, 43 and 44 volatile compounds were detected in T1, T2 and T3 samples, respectively. In addition, 9, 10 and 9 sulfur compounds were identified in three samples by GC-FPD, respectively. Combining physicochemical and sensory properties, T1 presented fatty, green and rotten cabbage odor; T2 exhibited mushroom, sulfuric and musty odor notes; T3 had nutty, floral and roasted potato odor. Dimethyl sulfide, 3-methylbutanal, dimethyl disulfide, 3-octanone, bis(methylthio) methane, octanal, 1-octen-3-one, 1-octen-3-ol and benzeneacetaldehyde played indispensable roles in the overall aroma of three truffles. Finally, based on quantitative concentration in T1, odorous compounds (OAV) > 1 were mixed to recombine aroma, demonstrating that these key aroma compounds based on OAV can successfully recombine pretty similar aroma of each variety.

Machine learning based classification of yogurt aroma types with flavoromics.

Traditional sensory evaluation, relying on human assessors, is vulnerable to subjective error and lacks automation. Nonetheless, the complexity of human sensation makes it challenging to develop a computational method in place of human sensory evaluation. To tackle this challenge, this study constructed logistic regression classification models that could predict yogurt aroma types based on aroma-active compound concentrations with high classification accuracy (AUC ROC > 0.8). Furthermore, indicator compounds discovered from feature importance analysis of classification models led to the derivation of classification criteria of yogurt aroma types. Through constructing and analyzing machine learning models on yogurt aroma types, this study provides an automated pipeline to monitor sensory properties of yogurts.

Integrated Analysis of Metabolomics, Flavoromics, and Transcriptomics for Evaluating New Varieties of Amomum villosum Lour.

Amomum villosum Lour. (A. villosum) is the original plant of the medicinal and culinary spice Amomi Fructus (Sharen) and is an important economic crop in the Lingnan region of China. During the cultivation and production of A. villosum, prolonged reliance on single asexual reproduction has exacerbated the degradation of its varieties, leading to inconsistent yields and quality. Building upon earlier cultivar selection efforts, this study provides a comprehensive evaluation of two newly bred A. villosum varieties (A11 and A12) from perspectives including plant traits, product characteristics, active ingredients, and multi-omics analysis. It was found that A12 plants display enhanced robustness, more aromatic fruits, higher yields, and elevated levels of bornyl acetate, A11 shows the advantage of a high camphor content, and the different metabolites and differentially expressed genes of the two varieties were significantly enriched in multiple metabolic pathways. Additionally, A12 contained more terpenoids and substances with aromatic odors such as sweet, fruity, floral, and green. Furthermore, a key gene (Wv_032842) regulating the acetylation of bornyl was discovered, and its significantly higher expression, in A12. In conclusion, this study has a guiding significance for the evaluation of germplasm resources and the breeding of excellent varieties of A. villosum.

Comparison of Aroma and Taste Profiles of Kiwi Wine Fermented with/without Peel by Combining Intelligent Sensory, Gas Chromatography-Mass Spectrometry, and Proton Nuclear Magnetic Resonance.

Kiwi wine (KW) is tipically made by fermenting juice from peeled kiwifruit, resulting in the disposal of peel and pomace as by-products. However, the peel contains various beneficial compounds, like phenols and flavonoids. Since the peel is edible and rich in these compounds, incorporating it into the fermentation process of KW presents a potential solution to minimize by-product waste. This study compared the aroma and taste profiles of KW from peeled (PKW) and unpeeled (UKW) kiwifruits by combining intelligent sensory technology, GC-MS, and 1H-NMR. Focusing on aroma profiles, 75 volatile organic compounds (VOCs) were identified in KW fermented with peel, and 73 VOCs in KW without peel, with 62 VOCs common to both. Among these compounds, rose oxide, D-citronellol, and bornylene were more abundant in UKW, while hexyl acetate, isoamyl acetate, and 2,4,5-trichlorobenzene were significantly higher in PKW. For taste profiles, E-tongue analysis revealed differences in the taste profiles of KW from the two sources. A total of 74 molecules were characterized using 1H-NMR. UKW exhibited significantly higher levels of tartrate, galactarate, N-acetylserotonin, 4-hydroxy-3-methoxymandelate, fumarate, and N-acetylglycine, along with a significantly lower level of oxypurinol compared to PKW. This study seeks to develop the theoretical understanding of the fermentation of kiwifruit with peel in sight of the utilization of the whole fruit for KW production, to increase the economic value of kiwifruit production.

Insights into the Flavor Profile of Yak Jerky from Different Muscles Based on Electronic Nose, Electronic Tongue, Gas Chromatography-Mass Spectrometry and Gas Chromatography-Ion Mobility Spectrometry.

It is well known that different muscles of yak exhibit distinctive characteristics, such as muscle fibers and metabolomic profiles. We hypothesized that different muscles could alter the flavor profile of yak jerky. Therefore, the objective of this study was to investigate the differences in flavor profiles of yak jerky produced by longissimus thoracis (LT), triceps brachii (TB) and biceps femoris (BF) through electronic nose (E-nose), electronic tongue (E-tongue), gas chromatography-mass spectrometry (GC-MS) and gas chromatography-ion mobility spectrometry (GC-IMS). The results indicated that different muscles played an important role on the flavor profile of yak jerky. And E-nose and E-tongue could effectively discriminate between yak jerky produced by LT, TB and BF from aroma and taste points of view, respectively. In particular, the LT group exhibited significantly higher response values for ANS (sweetness) and NMS (umami) compared to the BF and TB groups. A total of 65 and 47 volatile compounds were characterized in yak jerky by GC-MS and GC-IMS, respectively. A principal component analysis (PCA) model and robust principal component analysis (rPCA) model could effectively discriminate between the aroma profiles of the LT, TB and BF groups. Ten molecules could be considered potential markers for yak jerky produced by different muscles, filtered based on the criteria of relative odor activity values (ROAV) > 1, p < 0.05, and VIP > 1, namely 1-octen-3-ol, eucalyptol, isovaleraldehyde, 3-carene, D-limonene, γ-terpinene, hexanal-D, hexanal-M, 3-hydroxy-2-butanone-M and ethyl formate. Sensory evaluation demonstrated that the yak jerky produced by LT exhibited superior quality in comparison to that produced by BF and TB, mainly pertaining to lower levels of tenderness and higher color, taste and aroma levels. This study could help to understand the specific contribution of different muscles to the aroma profile of yak jerky and provide a scientific basis for improving the quality of yak jerky.

Effect of tea stems on the quality formation of large-leaf yellow tea: Sensomics and flavoromics approaches.

In this study, the stems (ST) and leaves (LT) isolated from Large-leaf yellow tea (LYT) were used for sensory evaluation and quantitative analysis of flavor metabolites by sensomics and flavoromics. The results showed that the flavors of ST and LT in LYT were significantly different, and ST had stronger roasty and nutty aroma and sweet taste, which was mainly due to the accumulation of higher theanine and soluble monosaccharides in ST, and provided more substrates for the production of more pyrazine by the Maillard reaction; whereas LT contributed to the mellow and thick taste quality of LYT, and the abundance of catechins and caffeine were the main reason. The metabolic patterns of flavor metabolites indicated that the flavor differences between ST and LT were mainly due to biological metabolism in tea plants. This study provides the selection of raw materials for LYT in the future and product development of tea stems.

Identification of Compounds That Contribute to Consumer Aroma Liking of Roasted American-European Hybrid Hazelnuts.

New interspecific hybrid hazelnut crosses between American (Corylus americana) and European (Corylus avellana) hazelnuts are being developed to support a commercial industry in the Midwest region of the United States. In this study, volatile compounds that impact consumer aroma liking of roasted hybrid hazelnuts (C. americana × C. avellana) were investigated by targeted and nontargeted GC/MS flavoromics. Chemical profiles from 10 roasted hybrid hazelnut samples were modeled with consumer aroma liking scores by orthogonal partial least-squares with good fit and predictive performance (R2 ≥ 0.92, Q2 ≥ 0.82, RMSECV = 0.2). Top ranked predictors positively correlated with liking included 12 aroma compounds and 4 profiled volatiles for the targeted and nontargeted methods, respectively. Sensory recombination testing of hazelnut samples with addition of the 12 predictive odorants was preferred by consumers (p < 0.001, Δ aroma liking = 2.2 on 9-point scale) and perceived as more roasty, nutty, and sweet compared to the control (p < 0.05). Addition of the 4 predictive volatiles at subthreshold levels also was preferred (p = 0.02) and perceived as less earthy and mushroom like than the control (p < 0.05).

The aroma profiles of dried gonggans: Characterization of volatile compounds in oven-dried and freeze-dried gonggan.

Dehydration is an effective method for the long-term storage and aroma retention of gonggan (Citrus sinensis Osb. 'Deqing Gonggan'), which is a Chinese variety of citrus, with unique and characteristic floral, fruity, and citrus flavors. However, the aroma profiles of gonggans prepared using oven- and freeze-drying, the most widely-used drying methods, remain unclear. In this study, a total of 911 volatile organic compounds (VOCs) were detected in dried gonggan. These were primarily composed of alcohols (7.69%), aldehydes (7.03%), esters (15.38%), ketones (7.58%), and terpenoids (23.19%). A total of 67 odorants contributed significantly to the overall aroma of dried gonggans, with the major odor qualities being detected as green, citrus, fruity, floral, and sweet. These were mainly attributed to the presence of aldehydes, esters, and terpenoids. Freeze-drying was more effective in maintaining the unique citrus and mandarin-like aromas attributed to compounds such as limonene, citrial, ß-myrcene, ß-pinene, and γ-terpinene. Moreover, (E,E)-2,4-decadienal had the highest relative odor activity value (rOAV) in freeze-dried gonggans, followed by (E)-2-nonenal, furaneol, (E, E)-2, 4-nonadienal, and E-2-undecenal. Oven-drying promoted the accumulation of terpenes such as octatriene, trans-ß-ocimene, cyclohexanone, copaene, and ɑ-irone, imparting a soft aroma of flowers, fruits, and sweet. Increasing the temperature led to an increase in existing VOCs or the generation of new VOCs through phenylpropanoid, terpenoid, and fatty acid metabolism. The findings of this study offer insights into an optimized procedure for producing high-quality dried gonggans. These insights can be valuable for the fruit-drying industry, particularly for enhancing the quality of dried fruits.

Flavoromics Analysis of Passion Fruit-Roasted Chicken.

Currently, research on the flavor components and their dynamic changes in roasted chicken with a special flavor is rare. In this study, a passion fruit-roasted chicken was prepared, its characteristic flavor components were profiled by flavoromics, and their evolution patterns and precursors were determined. The results showed that the characteristic flavor component with the highest contribution rate was ethyl butyrate (50.44%). In particular, some unique flavor compounds were identified compared with other roasted chicken products available. The main volatile flavor components in all stages of processing were alcohols, esters, and hydrocarbons, 15 to 30 min of roasting is an important stage for establishing the aroma system, and at the end, hydrocarbons were the main volatile compounds. During the 30-day storage period, the characteristic flavor components included ethyl butyrate, ethyl maltol, ß-caryophyllene, and guaiacene. In conclusion, passion fruit-roasted chicken contained many characteristic flavor components, which were mainly formed within 15 to 30 min of roasting and were basically stable during the 30-day storage period. In a word, this work prepared a novel roasted chicken and revealed its mechanism of flavor formation at different baking stages and storage periods, which provided references for industrial production.

Using in situ untargeted flavoromics analysis to unravel the empty cup aroma of Jiangxiang-type Baijiu: A novel strategy for geographical origin traceability.

"Empty cup aroma" is an important characteristic and quality evaluation standard of Jiangxiang-type Baijiu (JXB). In this study, an in situ detection method for the empty cup aroma of JXB was established, and the authenticity and origin information of JXB were identified with an untargeted flavoromics strategy. The complex composition of JXB leads to slow ethanol volatilization, which is a potential method for identifying artificial JXB. The results of the sensory analysis showed that acidic, sauce, burnt and qu in the empty cup of JXB were the strongest at the 45 min stage. A total of 155 compounds were detected in the empty cups of 15 JXB from different regions during 45 min of standing, and 34 compounds were identified as key aroma compounds in the empty cups of JXB. Eleven potential markers were screened (VIP > 1), which can be used to distinguish JXB produced in Guizhou/Sichuan and other regions.

Benchtop volatilomics supercharged: How machine learning based design of experiment helps optimizing untargeted GC-IMS gas phase metabolomics.

Gas chromatography-ion mobility spectrometry (GC-IMS) plays a significant role in both targeted and non-targeted analyses. However, the non-linear behavior of IMS and its complex ion chemistry pose challenges for finding optimal experimental conditions using existing methodologies. To address these issues, integrating machine learning (ML) strategies offers a promising approach. In this study, we propose a hybrid strategy, combining design of experiment (DOE) and machine learning (ML) for optimizing GC-IMS conditions in non-targeted volatilomic/flavoromic analysis, with saffron volatiles as a case study. To begin, a rotatable circumscribed central composite design (CCD) is used to define five influential GC-IMS factors of sample amount, headspace temperature, incubation time, injection volume, and split ratio. Subsequently, two ML models are utilized: multiple linear regression (MLR) as a linear model and Bayesian regularized-artificial neural network (BR-ANN) as a nonlinear model. These models are employed to predict the response variables of total peak areas (PAs) and the number of detected peaks (PNs) in GC-IMS. The findings show that there is a direct correlation between the factors in GC-IMS and the PNs, suggesting that MLR is a suitable approach for building a model in this scenario. However, the PAs exhibit nonlinear behavior, suggesting that BR-ANN is better suitable to capture this complexity. Notably, Derringer's desirability function is utilized to integrate the PAs and PNs, and in this scenario, MLR demonstrates satisfactory performance in modeling the GC-IMS factors.

Identification of Compounds That Impact Consumer Flavor Liking of American-European Hazelnut Hybrids Using Nontargeted LC/MS Analysis.

American-European (Corylus americana × Corylus avellana) hazelnut hybrids are being developed for the Midwest-growing region of the United States. However, an inadequate understanding of the compounds that impact the consumer acceptance of hazelnuts limits breeding programs. Nontargeted liquid chromatography/mass spectrometry (LC/MS) chemical profiles of 12 roasted hybrid hazelnut samples and the corresponding consumer flavor liking scores were modeled by orthogonal partial least squares with good fit and predictive ability (R2Y > 0.9, Q2 > 0.9) to identify compounds that impact nut liking. The five most predictive compounds (1-5) were negatively correlated to flavor liking, selected as putative markers, purified by multidimensional preparative LC/MS, structurally elucidated (nuclear magnetic resonance, MS), quantified, and validated for sensory relevance. Compound 1 was identified as 1″-O-3'-b-glucofuranosyl-1'-O-1-b-glucofuranosyl-(2,6-dihydroxyphenyl)-ethan-4-one. Compounds 2 and 4 were identified as rotamers of 2-(3-hydroxy-2-oxoindolin-3-yl) acetic acid 3-O-6'-galactopyranosyl-2″-(2″oxoindolin-3″yl) acetate, whereas compounds 3 and 5 were identified as rotamers of 1″-O-1'-b-glucofuranosyl-9-O-6'-b-glucopyranosyl-2″-(2″-oxoindolin-3″yl) acetate. Sensory evaluation determined that all compounds were characterized by bitterness and/or astringency. The sensory threshold values of compounds 1-5 were determined to be below the concentrations reported in 91, 83, 41, 25, and 41% of all 12 hybrid hazelnut samples, respectively, indicating they contributed to aversive flavor attributes.

Superior high-efficiency and high-throughput volatile flavor extraction of Japanese fermented seasonings by solvent-assisted stir bar solid extraction with reverse extraction.

Fermented seasonings have pleasant flavors that stimulate our appetite. Their flavoring properties change depending on factors such as their materials and fermented conditions. Therefore, a comparative analysis of their flavor is important when evaluating their quality. However, seasonings contain high levels of various matrices such as sugars, proteins, lipids, and ethanol, making it difficult to extract aroma compounds efficiently from them. In this study, we verified a high-efficient and high-throughput volatile flavor analysis of fermented seasonings by solvent-assisted stir bar solid extraction (SA-SBSE) with reverse extraction. We applied SA-SBSE to Japanese fermented seasonings, soy sauce, miso (fermented beans), and mirin (sweet rice wine) and compared their profiles with those from other common extraction methods, headspace gas-solid-phase microextraction (HS-SPME), liquid extraction with solvent-assisted flavor evaporation (LE-SAFE), and conventional SBSE (C-SBSE). The aroma properties and profiles of extracts from SA-SBSE were close to those of the original sample, being similar to that of LE-SAFE. In addition, potent aroma compounds in each sample were extracted by SA-SBSE and LE-SAFE, which were far superior to those by C-SBSE. For quantification, SA-SBSE extracts showed a good standard curve by the standard addition method. We could quantify maltol, one of the most common potent aroma compounds in all samples, for various commercial samples by such high-throughput analysis.

Unraveling the chemosensory characteristics on different types of spirits based on sensory contours and quantitative targeted flavoromics analysis.

Due to differences in raw materials and production processes, different spirits exhibit various flavor even if undergo distillation operation. In this study, sensory analysis could clearly distinguish 5 types spirits, and had been validated through quantitative targeted flavoromics analysis. Consequently, 44 potential differential markers between 5 types spirits were screened. Among, 34 definite differential markers were further confirmed to be highly correlated with target sensory attributes and could effectively distinguish types of spirits. Ultimately, 14 key differential markers (including 2-methylbutane, linalool, acetaldehyde, d-limonene, ß-myrcene, phenylethyl alcohol, phenethyl acetate, heptyl formate, ethyl octanoate, ethyl decanoate, ethyl pentanoate, ethyl hexanoate, hexanoic acid, and ethyl hexadecanoate) could reveal the chemical sources of spirit sensory and serve as targets for identifying different types of spirits. Overall, the results of flavoromic characterization of 5 types spirits provided a significant step forwards in understanding of differentiation of spirits by sensory coupled with quantitative, and statistical analysis.

A comparative study to determine the key aroma components of yogurt aroma types based on Sensomics and Flavoromics.

This study used Sensomics to examine four previously obtained yogurt aroma type profiles. 14 key aroma-active compounds were identified as significant contributors (p ≤ 0.05) in the four aroma types using gas chromatography-mass spectrometry-olfactometry (GC-MS/O), aroma extract dilution analysis (AEDA), odor activity values (OAV), and aroma recombination and omission experiments. The Sensomics and previous Flavoromics results were compared, showing that Flavoromics identified 10 indicator compounds for distinguishing aroma types. Eight were the same as the key aroma-active compounds identified via Sensomics, namely acetic acid, pentanoic acid, decanoic acid, 3-hydroxy-2-butanone, 2,3-pentanedione, acetaldehyde, δ-decalactone, and dimethyl sulfone. Sensomics revealed a prominent similarity between the categories of key aroma-active compounds of the four aroma types, with a higher sensory contribution. Flavoromics showed less overlapping between the indicator compounds, mainly related to the distinction between the four aroma types. Sensomics and Flavoromics serve distinct research objectives and should be selected according to the study subject.

The Effect of Imidacloprid on the Volatile Organic Compound Profile of Strawberries: New Insights from Flavoromics.

Organic agriculture is of great socioeconomic significance because it can promote the nutritional quality of horticultural crops and is environmentally friendly. However, owing to the lack of techniques for studying complex aroma-related chemical profiles, limited information is available on the influence of organic practices on the flavor quality of strawberries, one of the primary factors driving consumer preferences. Here, two-dimensional gas chromatography combined with time-of-flight mass spectrometry (GC×GC-TOF-MS) and flavoromics analysis was employed to investigate the profiles and differences in the volatile organic compounds (VOCs) of strawberries under organic (without imidacloprid) and conventional (with imidacloprid) agricultural practices. A total of 1164 VOCs, representing 23 chemical classes (e.g., aldehydes, terpenes, and furanone compounds), were detected, which is the highest number of VOCs that have ever been detected in strawberries. The sensory evaluation results indicated that there was a notable influence of imidacloprid (IMI) on the aroma of the strawberries. Principal component analysis and partial least squares discriminant analysis results suggested that the composition of volatile compounds significantly differed in the present study between the IMI-treated and non-IMI-treated groups. Furthermore, the flavor-related indicators of 25 key contributors to the differences between the two treatment groups suggested that VOC profiles can be considered an indicator for distinguishing between strawberries from different agricultural practices. Flavoromics can provide new insights into the quality of strawberries from different agricultural practices.

Effect of orange solid waste diet on flesh quality and metabolic profile of common carp (Cyprinus carpio).

In this study, we aimed to evaluate the effects of solid waste of Citrus sinensis (SWC) supplementation in diet on common carp (Cyprinus carpio) flesh quality and the potential mechanisms underlying these effects. Four diets, each with different levels of SWC (0%, 5%, 10%, and 15%), were formulated and administered to C. carpio (48.83 ± 5.59 g) for 60 days. The results showed that SWC diet significantly enhanced specific growth rate, muscle sweetness (via sweet amino acids and sweet molecules), and the nutritional value of fish meat (increased protein, α-vitamin E, and allopurinol). Chromatography-mass spectrometry analyses indicated that SWC supplementation increased the essential amino acid content in the diet. In addition, SWC diet promoted biosynthesis of non-essential amino acids in muscle by enhancing glycolysis and the tricarboxylic acid cycle. In conclusion, SWC could be a cost-effective solution for providing nutritious and flavourful aquatic products.

Identification of compounds that enhance bitterness of coffee brew.

The bitterness perception of coffee is a key attribute that impacts consumer acceptance. Nontargeted liquid chromatography/mass spectrometry (LC/MS) flavoromics analysis was applied to identify compounds that enhance the bitter perception of roasted coffee brew. Orthogonal partial least squares (OPLS) analysis was used to model the comprehensive chemical profiles and sensory bitter intensity ratings of fourteen coffee brews with good fit and predictivity. Five compounds that were highly predictive and positively correlated to bitter intensity were selected from the OPLS model, further isolated, and purified using preparative LC fractionation. Sensory recombination testing demonstrated that five compounds significantly enhanced the bitter perception of coffee when presented as a mixture, but not when presented individually. In addition, a set of roasting experiments revealed the five compounds were generated during the coffee roasting process.