Postfilter for Dual Channel Speech Enhancement Using Coherence and Statistical Model-Based Noise Estimation.

A multichannel speech enhancement system usually consists of spatial filters such as adaptive beamformers followed by postfilters, which suppress remaining noise. Accurate estimation of the power spectral density (PSD) of the residual noise is crucial for successful noise reduction in the postfilters. In this paper, we propose a postfilter utilizing proposed a posteriori speech presence probability (SPP) and noise PSD estimators, which are based on both the coherence and the statistical models. We model the coherence-based a posteriori SPP as a simple function of the magnitude of coherence between two microphone signals and combine it with a single-channel SPP based on statistical models. The coherence-based estimator for the PSD of the noise remaining in the beamformer output in the presence of speech is derived using the pseudo-coherence considering the effect of the beamformers, which is used to construct the coherence-based noise PSD estimator. Then, the final noise PSD estimator is obtained by combining the coherence-based and statistical model-based noise PSD estimators with the proposed SPP. The spectral gain function is also modified, incorporating the proposed SPP. Experimental results demonstrate that the proposed method led to more accurate noise PSD estimation and perceptual evaluation of speech quality scores in various diffuse noise environments, and did not degrade the speech quality under the presence of directional interference, although the proposed method utilizes the coherence information.

Sonoanatomy of the platysmal bands: What causes the platysmal band?

BACKGROUND: The platysmal band is created by the platysma muscle, a thin superficial muscle that covers the entire neck and the lower part of the face. The platysmal band appears at the anterior and posterior borders of the muscle. To date, no definite pathophysiology has been established. Here, we observed a lack of knowledge of the anatomy of the platysma muscle using ultrasonography in this study. METHODS: We conducted a descriptive, prospective study observing the platysmal band in resting and contraction states to reveal muscle changes. Twenty-four participants (aged 23-57 years) with anterior and posterior neck bands underwent ultrasonography in resting and contracted states. Ten cadavers were studied aged 67-85 years to measure the thickness of the platysma muscle at 12 points: horizontally (medial, middle, lateral) and vertically (inferior mandibular margin, hyoid bone, cricoid cartilage, superior margin of clavicle). RESULTS: The anterior and posterior borders of the platysma muscle were thicker than the middle of the platysma muscle when in a contracted state, and the muscle also had a convex shape when contracted. The thickness of the platysma muscle was not significantly different over 12 points in the resting state. During contraction, the platysma muscles contracted in the medial and lateral margins of the muscle, which was more significant in the posterior bands. CONCLUSION: The anterior and posterior platysmal bands are related to muscle thickness during contraction. These observations support the change in platysmal band treatment only at the anterior and posterior border of the muscle.

Comparison of Initial Presentation of Pediatric Diabetes Before and During the Coronavirus Disease 2019 Pandemic Era.

BACKGROUND: Hospital visitation has become challenging during the coronavirus disease 2019 pandemic because of quarantine measures and fear of infection. Consequently, newly diagnosed patients may present with more severe diseases during the pandemic. The present study analyzed the differences in the initial clinical presentations of newly diagnosed patients with type 1 diabetes (T1D) and type 2 diabetes (T2D), comparing pre-pandemic and pandemic periods. METHODS: Newly diagnosed patients with T1D or T2D and aged < 18 years during 2018-2020 were included in the study. Data were collected retrospectively from four academic centers in Gyeonggi-do, South Korea. Initial clinical data were compared between the pre-pandemic (2018-2019) and pandemic (2020) periods. RESULTS: In the pre-pandemic and pandemic periods, 99 patients (41 T1D and 58 T2D patients) and 84 patients (51 T1D and 33 T2D patients) were identified, respectively. During the pandemic, the proportion of diabetic ketoacidosis (DKA) cases increased compared to the pre-pandemic period (21.2% during 2018-2019 vs. 38.1% in 2020; P = 0.012). In the pre-pandemic and pandemic periods, initial pH was 7.32 ± 0.14 and 7.27 ± 0.15, respectively (P = 0.040), and HbA1c values were 11.18 ± 2.46% and 12.42 ± 2.87%, respectively (P = 0.002). During the pandemic, there was an increased risk of DKA in patients with T1D (odds ratio, 2.42; 95% confidence interval, 1.04-5.62; P = 0.040). CONCLUSION: During the pandemic, the proportion of DKA in newly diagnosed patients with T1D increased and clinical parameters showed a deteriorating pattern. Increased awareness of pediatric diabetes, especially DKA, could facilitate visit to the hospital for an early diagnosis; thus, reducing the number of DKA cases during the pandemic era.

Improved Speech Spatial Covariance Matrix Estimation for Online Multi-Microphone Speech Enhancement.

Online multi-microphone speech enhancement aims to extract target speech from multiple noisy inputs by exploiting the spatial information as well as the spectro-temporal characteristics with low latency. Acoustic parameters such as the acoustic transfer function and speech and noise spatial covariance matrices (SCMs) should be estimated in a causal manner to enable the online estimation of the clean speech spectra. In this paper, we propose an improved estimator for the speech SCM, which can be parameterized with the speech power spectral density (PSD) and relative transfer function (RTF). Specifically, we adopt the temporal cepstrum smoothing (TCS) scheme to estimate the speech PSD, which is conventionally estimated with temporal smoothing. Furthermore, we propose a novel RTF estimator based on a time difference of arrival (TDoA) estimate obtained by the cross-correlation method. Furthermore, we propose refining the initial estimate of speech SCM by utilizing the estimates for the clean speech spectrum and clean speech power spectrum. The proposed approach showed superior performance in terms of the perceptual evaluation of speech quality (PESQ) scores, extended short-time objective intelligibility (eSTOI), and scale-invariant signal-to-distortion ratio (SISDR) in our experiments on the CHiME-4 database.

Understanding the Formation and Mechanism of Anticipatory Responses in Escherichia coli.

Microorganisms often live in complex habitats, where changes in the environment are predictable, providing an opportunity for microorganisms to learn, anticipate the upcoming environmental changes and prepare in advance for better survival and growth. One such environment is the mammalian intestine, where the abundance of different carbon sources is spatially distributed. In this study, we identified seven spatially distributed carbon sources in the mammalian intestine and tested whether Escherichia coli exhibits phenotypes that are consistent with an anticipatory response given their spatial order and abundance within the mammalian intestine. Through RNA-Seq and RT-PCR validation measurements, we found that there was a 67% match in the expression patterns between the measured phenotypes and what would otherwise be expected in the case of anticipatory behavior, while 83% and 0% were in agreement with the homeostatic and random response, respectively. To understand the genetic and phenotypic basis of the discrepancies between the expected and measured anticipatory responses, we thoroughly investigated the discrepancy in D-galactose treatment and the expression of maltose operon in E. coli. Here, the expected anticipatory response, based on the spatial distribution of D-galactose and D-maltose, was that D-galactose should upregulate the maltose operon, but it was the opposite in experimental validation. We performed whole genome random mutagenesis and screening and identified E. coli strains with positive expression of maltose operon in D-galactose. Targeted Sanger sequencing and mutation repair identified that the mutations in the promoter region of malT and in the coding region of the crp gene were the factors responsible for the reversion in the association. Further, to identify why positive association in the D-galactose treatment and the expression of the maltose operon did not evolve naturally, fitness measurements were performed. Fitness experiments demonstrated that the fitness of E. coli strains with a positive association in the D-galactose treatment and the expression of the maltose operon was 12% to 20% lower than that of the wild type strain.

Compendium of synovial signatures identifies pathologic characteristics for predicting treatment response in rheumatoid arthritis patients.

Rheumatoid arthritis (RA) is therapeutically challenging due to patient heterogeneity and variability. Herein we describe a novel integration of RA synovial genome-scale transcriptomic profiling of different patient cohorts that can be used to provide predictive insights on drug responses. A normalized compendium consisting of 256 RA synovial samples that cover an intersection of 11,769 genes from 11 datasets was build and compared with similar datasets derived from OA patients and healthy controls. Differentially expression genes (DEGs) that were identified in three independent methods were fed into functional network analysis, with subsequent grouping of the samples based on a non-negative matrix factorization method. RA-relevant pathway activation scores and four machine learning classification techniques supported the generation of a predictive model of patient treatment response. We identified 876 up-regulated DEGs including 24 known genetic risk factors and 8 drug targets. DEG-based subgrouping revealed 3 distinct RA patient clusters with distinct activity signatures for RA-relevant pathways. In the case of infliximab, we constructed a classifier of drug response that was highly accurate with an AUC/AUPR of 0.92/0.86. The most informative pathways in achieving this performance were the NFκB-, FcεRI- TCR-, and TNF signaling pathways. Similarly, the expression of the HMMR, PRPF4B, EVI2A, RAB27A, MALT1, SNX6, and IFIH1 genes contributed in predicting the patient outcome. Construction and analysis of normalized synovial transcriptomic compendia can provide useful insights for understanding RA-related pathway involvement and drug responses for individual patients.

Population collapse and adaptive rescue during long-term chemostat fermentation.

Microbial fermentation is an essential process for research and industrial applications, yet our understanding of cellular dynamics during long-term fermentation is limited. Here, we report a reproducible phenomenon of abrupt population collapse followed by a rapid population rescue that was observed during long-term chemostat cultivations, for various strains of Escherichia coli in minimal media. Through genome resequencing and whole-genome transcriptional profiling of replicate runs over time, we identified that changes in the tRNA and carbon catabolic genes are the genetic basis of this phenomenon. Since current fermentation models are unable to capture the observed dynamics, we present an extended model that takes into account critical biological processes during fermentation, and we further validated carbon source predictions through forward experimentation. This study extends the predictability of current models for microbial fermentation and adds to our system-level knowledge of cellular adaptation during this crucial biotechnological process.

The Genetic and Transcriptional Basis of Short and Long Term Adaptation across Multiple Stresses in Escherichia coli.

Microbes exhibit short and long term responses when exposed to challenging environmental conditions. To what extent these responses are correlated, what their evolutionary potential is and how they translate to cross-stress fitness is still unclear. In this study, we comprehensively characterized the response of Escherichia coli populations to four abiotic stresses (n-butanol, osmotic, acidic, and oxidative) and their combinations by performing genome-scale transcriptional analysis and growth profiling. We performed an analysis of their cross-stress behavior which identified 15 cases of cross- protection and one case of cross vulnerability. To elucidate the evolutionary potential of stress responses to individual stresses and stress combinations, we re-sequenced E. coli populations evolved in those four environments for 500 generations. We developed and applied a network-driven method that integrates mutations and differential expression to identify core and stress-specific gene communities that are likely to have a phenotypic impact. Our results suggest that beyond what is expected from the general stress response mechanisms, cross-stress behavior arises both from common pathways, several including metal ion binding and glycolysis/gluconeogenesis, and stress-specific expression programs. The stress-specific dependences uncovered, argue that cross-stress behavior is ubiquitous and central to understanding microbial physiology under stressful conditions.

DeepPep: Deep proteome inference from peptide profiles.

Protein inference, the identification of the protein set that is the origin of a given peptide profile, is a fundamental challenge in proteomics. We present DeepPep, a deep-convolutional neural network framework that predicts the protein set from a proteomics mixture, given the sequence universe of possible proteins and a target peptide profile. In its core, DeepPep quantifies the change in probabilistic score of peptide-spectrum matches in the presence or absence of a specific protein, hence selecting as candidate proteins with the largest impact to the peptide profile. Application of the method across datasets argues for its competitive predictive ability (AUC of 0.80±0.18, AUPR of 0.84±0.28) in inferring proteins without need of peptide detectability on which the most competitive methods rely. We find that the convolutional neural network architecture outperforms the traditional artificial neural network architectures without convolution layers in protein inference. We expect that similar deep learning architectures that allow learning nonlinear patterns can be further extended to problems in metagenome profiling and cell type inference. The source code of DeepPep and the benchmark datasets used in this study are available at https://deeppep.github.io/DeepPep/.

Empirical prediction of genomic susceptibilities for multiple cancer classes.

An empirical approach is presented for predicting the genomic susceptibility of an individual to the most likely one among nine traits, consisting of eight major cancer classes plus a healthy trait. We use four prediction methods by applying two supervised learning algorithms to two different descriptors of common genomic variations (the profiles of genotypes of SNPs and SNP syntaxes with low P values or low frequencies) of each individual genome from normal cells. All four methods made correct predictions substantially better than random predictions for most cancer classes, but not for some others. A combination of the four results using Bayesian inference better predicted overall than any individual method. The multiclass accuracy of the combined prediction ranges from 33% to 56% depending on cancer classes of testing sets, compared with 11% for a random prediction among nine traits. Despite limited SNP data available and the absence of rare SNPs in public databases, at present, the results suggest that the framework of this approach or its improvement can predict cancer susceptibility with probability estimates useful for making health decisions for individuals or for a population.

Microbial forensics: predicting phenotypic characteristics and environmental conditions from large-scale gene expression profiles.

A tantalizing question in cellular physiology is whether the cellular state and environmental conditions can be inferred by the expression signature of an organism. To investigate this relationship, we created an extensive normalized gene expression compendium for the bacterium Escherichia coli that was further enriched with meta-information through an iterative learning procedure. We then constructed an ensemble method to predict environmental and cellular state, including strain, growth phase, medium, oxygen level, antibiotic and carbon source presence. Results show that gene expression is an excellent predictor of environmental structure, with multi-class ensemble models achieving balanced accuracy between 70.0% (±3.5%) to 98.3% (±2.3%) for the various characteristics. Interestingly, this performance can be significantly boosted when environmental and strain characteristics are simultaneously considered, as a composite classifier that captures the inter-dependencies of three characteristics (medium, phase and strain) achieved 10.6% (±1.0%) higher performance than any individual models. Contrary to expectations, only 59% of the top informative genes were also identified as differentially expressed under the respective conditions. Functional analysis of the respective genetic signatures implicates a wide spectrum of Gene Ontology terms and KEGG pathways with condition-specific information content, including iron transport, transferases, and enterobactin synthesis. Further experimental phenotypic-to-genotypic mapping that we conducted for knock-out mutants argues for the information content of top-ranked genes. This work demonstrates the degree at which genome-scale transcriptional information can be predictive of latent, heterogeneous and seemingly disparate phenotypic and environmental characteristics, with far-reaching applications.

The Effects of Long-Term Ketamine Treatment on Cognitive Function in Complex Regional Pain Syndrome: A Preliminary Study.

OBJECTIVE: This preliminary study aimed to investigate the effects of long-term frequent ketamine treatment on cognitive function in [AQ-A] CRPS patients. DESIGN: A total of 30 CRPS patients were divided into two groups based on both the duration and frequency of ketamine treatment; the long-term frequent ketamine treatment (LF) group (N = 14) and the Non-LF group (N = 16). Participants were asked to complete a questionnaire packet including demographic and clinical characteristics and potential variables affecting cognitive function. Then, they performed the neuropsychological test. RESULTS: Results indicated that the LF group performed significantly poorer than the Non-LF group on the digit span, digit symbol, Controlled Oral Word Association Test, and Trail Making Test, but not the Stroop task. CONCLUSIONS: Patients with CRPS receiving long-term frequent ketamine treatment showed impairment in cognitive function (specifically executive function) compared with those who do not. These findings may have implications for clinical assessment and rehabilitation of cognitive function in CRPS patients.

Increased power of mixed models facilitates association mapping of 10 loci for metabolic traits in an isolated population.

The potential benefits of using population isolates in genetic mapping, such as reduced genetic, phenotypic and environmental heterogeneity, are offset by the challenges posed by the large amounts of direct and cryptic relatedness in these populations confounding basic assumptions of independence. We have evaluated four representative specialized methods for association testing in the presence of relatedness; (i) within-family (ii) within- and between-family and (iii) mixed-models methods, using simulated traits for 2906 subjects with known genome-wide genotype data from an extremely isolated population, the Island of Kosrae, Federated States of Micronesia. We report that mixed models optimally extract association information from such samples, demonstrating 88% power to rank the true variant as among the top 10 genome-wide with 56% achieving genome-wide significance, a >80% improvement over the other methods, and demonstrate that population isolates have similar power to non-isolate populations for observing variants of known effects. We then used the mixed-model method to reanalyze data for 17 published phenotypes relating to metabolic traits and electrocardiographic measures, along with another 8 previously unreported. We replicate nine genome-wide significant associations with known loci of plasma cholesterol, high-density lipoprotein, low-density lipoprotein, triglycerides, thyroid stimulating hormone, homocysteine, C-reactive protein and uric acid, with only one detected in the previous analysis of the same traits. Further, we leveraged shared identity-by-descent genetic segments in the region of the uric acid locus to fine-map the signal, refining the known locus by a factor of 4. Finally, we report a novel associations for height (rs17629022, P< 2.1 × 10⁻8).

Cardiometabolic risk factors and metabolic syndrome based on severity of obesity in Korean children and adolescents: data from the Korea National Health and Nutrition Examination Survey 2007-2018.

PURPOSE: Data regarding cardiometabolic risk factors (CMRFs) and metabolic syndrome (MetS) by body mass index (BMI) category in Korean youth are sparse. METHODS: Among the participants of the Korea National Health and Nutrition Examination Survey 2007-2018, 9,984 youth aged 10-18 years were included in the study. Participants were classified into 4 groups based on BMI status: normal weight, overweight, class I, and class II/III obesity. CMRF prevalence, including total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, triglycerides, blood pressure, fasting glucose, glycated hemoglobin, and MetS, were determined using the International Diabetes Federation (IDF) and modified National Cholesterol Education Program-Adult Treatment Panel (NCEP-ATP) III criteria based on BMI category. RESULTS: The prevalence of overweight, class I, class II, and class III obesity was 9.52%, 7.73%, 2.10%, and 0.32%, respectively. Mean CMRF values increased with BMI, except high-density lipoprotein cholesterol. Age- and sex-adjusted odds ratios (ORs) for prediction of CMRFs also increased with BMI. Adjusted ORs for MetS among overweight, class I, and class II/II obesity were 54.2, 283.3, and 950.3 for IDF criteria and 9.56, 37.9, and 126.8 for NCEP-ATP III criteria, respectively (all p<0.001). CONCLUSION: Class II and III obesity in Korean children and adolescents was associated with significantly increased CMRF and MetS prevalence. Therefore, it can be useful to measure CMRFs in obese children and adolescents. Further studies are required to establish screening guidelines based on obesity severity.

A Novel Approach on Deep Learning-Based Decision Support System Applying Multiple Output LSTM-Autoencoder: Focusing on Identifying Variations by PHSMs' Effect over COVID-19 Pandemic.

Following the outbreak of the COVID-19 pandemic, the continued emergence of major variant viruses has caused enormous damage worldwide by generating social and economic ripple effects, and the importance of PHSMs (Public Health and Social Measures) is being highlighted to cope with this severe situation. Accordingly, there has also been an increase in research related to a decision support system based on simulation approaches used as a basis for PHSMs. However, previous studies showed limitations impeding utilization as a decision support system for policy establishment and implementation, such as the failure to reflect changes in the effectiveness of PHSMs and the restriction to short-term forecasts. Therefore, this study proposes an LSTM-Autoencoder-based decision support system for establishing and implementing PHSMs. To overcome the limitations of existing studies, the proposed decision support system used a methodology for predicting the number of daily confirmed cases over multiple periods based on multiple output strategies and a methodology for rapidly identifying varies in policy effects based on anomaly detection. It was confirmed that the proposed decision support system demonstrated excellent performance compared to models used for time series analysis such as statistical models and deep learning models. In addition, we endeavored to increase the usability of the proposed decision support system by suggesting a transfer learning-based methodology that can efficiently reflect variations in policy effects. Finally, the decision support system proposed in this study provides a methodology that provides multi-period forecasts, identifying variations in policy effects, and efficiently reflects the effects of variation policies. It was intended to provide reasonable and realistic information for the establishment and implementation of PHSMs and, through this, to yield information expected to be highly useful, which had not been provided in the decision support systems presented in previous studies.

A functional single nucleotide polymorphism at the promoter region of cyclin A2 is associated with increased risk of colon, liver, and lung cancers.

BACKGROUND: The objective of this was to identify functional single nucleotide polymorphisms (SNPs) in cyclin-dependent kinases (CDKs) and cyclins that are associated with risk of human cancer. METHODS: First, 45 SNPs in CDKs and cyclins were analyzed in 106 lung cancers and 108 controls for a pilot study. One SNP (reference SNP [rs] 769236, +1 guanine to adenine [G→A]) at the promoter region of cyclin A2 (CCNA2) also was analyzed in 1989 cancers (300 breast cancers, 450 colorectal cancers, 450 gastric cancers, 367 hepatocellular carcinomas, and 422 lung cancers) and in 1096 controls. Genotyping was performed using matrix-assisted laser desorption-ionization/time-of-flight mass spectrometry. Transcriptional activity of the SNP according to the cell cycle was analyzed by using a luciferase reporter assay and fluorescence-activated cell sorting analysis in NIH3T3 cells. RESULTS: In the pilot study, the SNP (rs769236) was associated significantly with the risk of lung cancer. In the expanded study, multivariate logistic regression indicated that the AA homozygous variant of the SNP was associated significantly with the development of lung cancer (P < .0001; codominant model), colorectal cancer (P < .0001), and hepatocellular carcinoma (P = .02) but not with breast cancer or gastric cancer. The luciferase activity of a 300-base pair construct that contained the A allele was 1.5-fold greater than the activity of a construct with the G allele in NIH3T3 cells. The high luciferase activity of constructs that contained the A allele did not change with cell cycle progression. CONCLUSIONS: The current results suggested that an SNP (rs769236) at the promoter of CCNA2 may be associated significantly with increased risk of colon, liver, and lung cancers.

The Computational Diet: A Review of Computational Methods Across Diet, Microbiome, and Health.

Food and human health are inextricably linked. As such, revolutionary impacts on health have been derived from advances in the production and distribution of food relating to food safety and fortification with micronutrients. During the past two decades, it has become apparent that the human microbiome has the potential to modulate health, including in ways that may be related to diet and the composition of specific foods. Despite the excitement and potential surrounding this area, the complexity of the gut microbiome, the chemical composition of food, and their interplay in situ remains a daunting task to fully understand. However, recent advances in high-throughput sequencing, metabolomics profiling, compositional analysis of food, and the emergence of electronic health records provide new sources of data that can contribute to addressing this challenge. Computational science will play an essential role in this effort as it will provide the foundation to integrate these data layers and derive insights capable of revealing and understanding the complex interactions between diet, gut microbiome, and health. Here, we review the current knowledge on diet-health-gut microbiota, relevant data sources, bioinformatics tools, machine learning capabilities, as well as the intellectual property and legislative regulatory landscape. We provide guidance on employing machine learning and data analytics, identify gaps in current methods, and describe new scenarios to be unlocked in the next few years in the context of current knowledge.

Predicting early risk of chronic kidney disease in cats using routine clinical laboratory tests and machine learning.

BACKGROUND: Advanced machine learning methods combined with large sets of health screening data provide opportunities for diagnostic value in human and veterinary medicine. HYPOTHESIS/OBJECTIVES: To derive a model to predict the risk of cats developing chronic kidney disease (CKD) using data from electronic health records (EHRs) collected during routine veterinary practice. ANIMALS: A total of 106 251 cats that attended Banfield Pet Hospitals between January 1, 1995, and December 31, 2017. METHODS: Longitudinal EHRs from Banfield Pet Hospitals were extracted and randomly split into 2 parts. The first 67% of the data were used to build a prediction model, which included feature selection and identification of the optimal neural network type and architecture. The remaining unseen EHRs were used to evaluate the model performance. RESULTS: The final model was a recurrent neural network (RNN) with 4 features (creatinine, blood urea nitrogen, urine specific gravity, and age). When predicting CKD near the point of diagnosis, the model displayed a sensitivity of 90.7% and a specificity of 98.9%. Model sensitivity decreased when predicting the risk of CKD with a longer horizon, having 63.0% sensitivity 1 year before diagnosis and 44.2% 2 years before diagnosis, but with specificity remaining around 99%. CONCLUSIONS AND CLINICAL IMPORTANCE: The use of models based on machine learning can support veterinary decision making by improving early identification of CKD.

Data integration and predictive modeling methods for multi-omics datasets.

Translating data to knowledge and actionable insights is the Holy Grail for many scientific fields, including biology. The unprecedented massive and heterogeneous data have created as many challenges to store, process and analyze as the opportunities and promises they hold. Here, we provide an overview of these opportunities and challenges in multi-omics predictive analytics.

Predicting the evolution of Escherichia coli by a data-driven approach.

A tantalizing question in evolutionary biology is whether evolution can be predicted from past experiences. To address this question, we created a coherent compendium of more than 15,000 mutation events for the bacterium Escherichia coli under 178 distinct environmental settings. Compendium analysis provides a comprehensive view of the explored environments, mutation hotspots and mutation co-occurrence. While the mutations shared across all replicates decrease with the number of replicates, our results argue that the pairwise overlapping ratio remains the same, regardless of the number of replicates. An ensemble of predictors trained on the mutation compendium and tested in forward validation over 35 evolution replicates achieves a 49.2 ± 5.8% (mean ± std) precision and 34.5 ± 5.7% recall in predicting mutation targets. This work demonstrates how integrated datasets can be harnessed to create predictive models of evolution at a gene level and elucidate the effect of evolutionary processes in well-defined environments.