Machine learning for metabolic engineering: A review

Metabolic Engineering, Journal Year: 2020, Volume and Issue: 63, P. 34 - 60

Published: Nov. 20, 2020

Machine learning provides researchers a unique opportunity to make metabolic engineering more predictable. In this review, we offer an introduction discipline in terms that are relatable engineers, as well providing in-depth illustrative examples leveraging omics data and improving production. We also include practical advice for the practitioner of management, algorithm libraries, computational resources, important non-technical issues. A variety applications ranging from pathway construction optimization, genetic editing cell factory testing, production scale-up discussed. Moreover, promising relationship between machine mechanistic models is thoroughly reviewed. Finally, future perspectives most directions combination disciplines examined.

Language: Английский

---

Variability of plasmid fitness effects contributes to plasmid persistence in bacterial communities

Nature Communications, Journal Year: 2021, Volume and Issue: 12(1)

Published: May 11, 2021

Abstract Plasmid persistence in bacterial populations is strongly influenced by the fitness effects associated with plasmid carriage. However, wild-type hosts remain largely unexplored. In this study, we determined of major antibiotic resistance pOXA-48_K8 wild-type, ecologically compatible enterobacterial isolates from human gut microbiota. Our results show that although produced an overall reduction fitness, it small most hosts, and even beneficial several isolates. Moreover, genomic showed a link between phylogeny, helping to explain epidemiology. Incorporating our into simple population dynamics model revealed new set conditions for stability communities, increasing diversity becoming less dependent on conjugation. These help high prevalence plasmids greatly diverse natural microbial communities.

Language: Английский

---

Engineered Living Hydrogels

Advanced Materials, Journal Year: 2022, Volume and Issue: 34(26)

Published: March 4, 2022

Living biological systems, ranging from single cells to whole organisms, can sense, process information, and actuate in response changing environmental conditions. Inspired by living engineered nonliving matrices are brought together, which gives rise the technology of materials. By designing functionalities structures matrices, materials be created detect variability surrounding environment adjust their functions accordingly, thereby enabling applications health monitoring, disease treatment, remediation. Hydrogels, a class soft, wet, biocompatible materials, have been widely used as for cells, leading nascent field hydrogels. Here, interactions between hydrogel described, focusing on how hydrogels influence cell behaviors affect properties. The environments, these enable versatile applications, also discussed. Finally, current challenges facing clinical settings highlighted.

Language: Английский

---

Digital health: trends, opportunities and challenges in medical devices, pharma and bio-technology

CSI Transactions on ICT, Journal Year: 2023, Volume and Issue: 11(1), P. 11 - 30

Published: April 1, 2023

Language: Английский

---

Treatment options of nitrogen heterocyclic compounds in industrial wastewater: from fundamental technologies to energy valorization applications and future process design strategies

Water Research, Journal Year: 2025, Volume and Issue: unknown, P. 123575 - 123575

Published: March 1, 2025

Language: Английский

---

Observation of universal ageing dynamics in antibiotic persistence

Nature, Journal Year: 2021, Volume and Issue: 600(7888), P. 290 - 294

Published: Nov. 17, 2021

Language: Английский

---

Conjugation dynamics depend on both the plasmid acquisition cost and the fitness cost

Molecular Systems Biology, Journal Year: 2021, Volume and Issue: 17(3)

Published: March 1, 2021

Article1 March 2021Open Access Transparent processSource Data Conjugation dynamics depend on both the plasmid acquisition cost and fitness Hannah Prensky Department of Biology, Barnard College, New York, NY, USA Search for more papers by this author Angela Gomez-Simmonds Division Infectious Diseases, Medicine, Columbia University Irving Medical Center, Anne-Catrin Uhlemann Allison J Lopatkin Corresponding Author [email protected] orcid.org/0000-0003-0018-9205 Ecology, Evolution, Environmental University, Science Institute, Information Prensky1, Gomez-Simmonds2, Uhlemann2 *,1,3,4 1Department 2Division 3Department 4Data *Corresponding author. Tel: +1 212 853 2564; E-mail: Molecular Systems Biology (2021)17:e9913https://doi.org/10.15252/msb.20209913 PDFDownload PDF article text main figures. Peer ReviewDownload a summary editorial decision process including letters, reviewer comments responses to feedback. ToolsAdd favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Plasmid conjugation is major mechanism responsible spread antibiotic resistance. costs are known impact long-term growth microbial populations providing plasmid-carrying cells relative (dis)advantage compared plasmid-free counterparts. Separately, introduces an immediate, but transient, metabolic perturbation. However, these short-term effects subsequent has not previously been established. Here, we observed that de novo transconjugants grew significantly slower and/or with overall prolonged lag times, lineages had replicating several generations, indicating presence cost. These were general diverse incompatibility groups, well-characterized clinically captured plasmids, Gram-negative recipient strains species, experimental conditions. Modeling revealed modulate dynamics, validated published data. results suggest hours immediately following may play critical role in short- prevalence. This time frame particularly relevant microbiomes high plasmid/strain diversity considered be hot spots conjugation. Synopsis Quantification shows indicates short A novel framework quantifies independently effects. The magnitude potentially dictated initial energetic burden imposed newly acquired plasmid, as well host cells' ability accommodate given environment. Incorporating into mathematical model improves temporal predictions dynamics. window represent interval quantifying Introduction Horizontal gene transfer (HGT) conjugation, which refers DNA from donor through direct cell-to-cell contact (Frost Koraimann, 2010) (Fig 1A), predominant way bacteria mobilize exchange resistance genes (Maiden, 1998; Barlow, 2009). can occur via chromosomally integrated conjugative elements or autonomously plasmids. Plasmids, often encode one multiple (Holmes et al, 2016), primarily global dissemination since approximately half all plasmids (Smillie have broad ranges (Klumper 2015). Moreover, postulated prevalent complex communities (e.g., gut soil microbiomes) due local density, diversity, abundance strains/species along mobile genetic (Ogilvie 2012). Figure 1. Transconjugants exhibit Schematic whereby (D, blue) transferred (R, red), generating transconjugant (T, yellow). R D each resistant (A1 A2, respectively), sensitive other. (T) uniquely antibiotics. protocol involves mixing R, followed one-hour incubation at 25°C. Cells then diluted media containing A1 A2 over 96-well plate. OD600 T (aqua) adapted (gray), (RP4 transconjugants) initiated same number per well. Each curve biological replicate. Black dashed lines best-fits. Growth rate (left) (right) T, after 1 h re-grown 24 h. De rates statistically less than (gray) (P = 1.12e-08, Appendix Table S2) (red) 7.27e-09, S2). times greater 5.71e-08 (3.77e-09, replicates. All statistics done using one-way ANOVA Bonferroni correction. Left: 7.34e-05 4.95e-05 15 120 min, respectively). Right: Lag multiplied true T0 divided mean (Appendix Fig S1D non-normalized). normalized 1.30e-09 1.31e-09 After h, condition (E) was re-grown. identical 1.00 min). 0.69 P 0.48 described (E). data Source available online figure. [msb20209913-sup-0002-SDataFig1.zip] Download figure PowerPoint formation proliferation new HGT progeny (termed fundamentally governed two kinetic processes: efficiency) dynamics). Both efficiency extrinsic intrinsic factors. For example, cell's physiological state drastically alter certain orders (Lopatkin 2016b). Likewise, highly dependent functional benefit yield net positive negative rates. Measuring processes enabled accurate persistence simple, vitro communities, consisting few strains, thereby improving our understanding 2017). native bacterial consist hundreds species interact, grow, compete scales (Jorgensen 2014). Predicting fate settings remains challenge (Dunn 2019; Collins, 2020). It widely that, absence selection, exert their hosts. Fitness vary attributed plasmid. plasmid's typically quantified established lineage isolated clone), competition assays with, to, counterpart (Ponciano 2007). Previous studies shown population structure dynamics; costly out-competed, leading elimination, compensated mutations ameliorate burden, prolonging (Dahlberg Chao, 2003; Dionisio, 2005; Harrison 2015; Loftie-Eaton Separate costs, acquiring requires immediate adaptation altered regulation resource allocation (San Millan 2018)) therefore also impacts cellular metabolism. it plasmid-encoded stress response transiently expressed 20–40 min (Althorpe 1999; Baharoglu 2010); SOS accounts considerable component maintenance metabolism (Kempes recent work demonstrated overshoot expression only occurs recently generated (Fernandez-Lopez extent disruption suggests cost, considerably focused effects, refer Indeed, traditional measurements they soon quickly stabilize la Cruz, 2014); manifest such preceding exponential growth, renders quantification challenging. Overall, generality remain unknown. Given natural environments, shed insights fates mixed/competing populations. discovered exhibited transient reduced increased conjugation—a corresponding establishing efficiently fulfill incorporating improved across range demonstrate prevalence importance implications predicting dominant Results Acquisition RP4 To investigate how might affect sought compare (which undergone adaptation) those (and fully adapted). readily estimating efficiencies 2016a): donors (D) recipients (R) carrying unique mixed under conditions minimize growth. Since antibiotics, directly selected population, its tracked microplate reader 1B). procedure ideal purposes minimizes during window, ensuring dynamic characterization captures emergent phenotypic changes. In contrast, streaking mixtures onto dual agar plates; individual clones grown stored testing Rozwandowicz 2019). stably reproducible used quantify determine timescale compensatory (Harrison Hall Using approach, first well-established, large conjugal (~60 kb). We chose imposing cell, allows us distinguish between thereafter S1A). Briefly, Escherichia coli MG1655 S1A); expresses spectinomycin (Spec) resistance, D, carries encodes kanamycin (Kan) measure hour 25°C, 1,000× Spec-Kan liquid media, reader. parallel, incubated hour, 25°C) control any itself, subsequently comparable cell verified colony forming units (CFU) S1B). Strikingly, appeared grow 1C). curve-fitting Baranyi equation (Baranyi Roberts, 1994) T's lower, higher, 1D, S1C, 1.12e-08 5.71e-09, independent duration: similar trends 1E S1C). when restored 1F). Importantly, cases, recovered remained lower strain, retained 1D). Finally, method parameters three additional methods resulted qualitatively consistent S2, Although initially costly, identified protocol-related factors could account observations. First, adjust environment, possibly altering conjugation-specific subjected conditions, effect environmental Second, residual R/D test this, background media; approximates parental densities present experiment. Doing so did trajectory nor discrepancy S1D). there no appreciable density S1E), neither survived long enough conjugate window. conclude indeed acquire, time. Introducing quantitative metric That induced changes intuitive: interpretations associated replication/protein expression. perturbation nutrient shift (Madar 2013). Therefore, facilitate quantification, define rigorous would capture end, minimal suggested required reach predetermined threshold inclusive proxy 2A S3); study analogous "time threshold" (Bethke 2. (t*, orange) specified (T*, top purple) reached depends (T0, bottom purple), (µ, aqua) (λ, orange). Assuming subtraction, line represented shown. Representative standard generation generate curve, 10-fold increments measured (dark light gray T0). Circles indicate t* (purple line) T* (orange line). Aqua represents out-growth experiment (i.e., T). plotting against t*; black curve. decreasing gray); blue markers show 0.275 (t*). Standard blue. Error bars deviation True predicted CFU E. strain MG1655. Scatter points replicates, bar height average. K. pneumoniae (KPN) AL2425. four 2 [msb20209913-sup-0003-SDataFig2.zip] let T(t) describe time, µ maximum specific rate. Consistent previous literature (Métris 2006), extrapolating region horizontal axis geometric (λ), corresponds observable onset 2A). During phase, thus line: ln(T(t)) ln(T0) + µ(t-λ), where density. Under (t*) takes detection level (T*) inversely correlated T0. other words, predict unknown (Tpred) value, assuming T*, µ, λ constant 2B). Conversely, Tpred T0, determine, specifically changed. purposes, consider relating above. case, Tpred/T0 < 1) growth-specific consequences use rather t*, simultaneously Additionally, priori spanning avoids variability inherent manually diluting target number, As such, robust trustworthy method. utility confirming discrepancies Specifically, built 2C) based quantitated curves. Having counts, found 0.0143, right-tailed t-test), 2D), expected. note approximate does change Figs S4A B), earlier. Generality determined whether particular estimates different dilution factor, strain). S4C i), ii), factor iii-iv); indeed, 150×, 500×, 1,000×, 5,000× separate parallel S4D), systematic difference estimates. acquire Klebsiella pneumoniae(KPN), species-level 2E). drastic KPN solely function plasmids; strain/species-level attributes likely key Next, reasoned replication kb), coupled significant amount energy synthesize machinery upon acquisition, immense led inhibition, thus, MacLean, hypothesized mobilizable trans do themselves machinery, most result encoding reduces size production. FHR helper system (Dimitriu 2014), self-transmissible co-residing recognition sequence oriT. pR chloramphenicol RP4, exhibits S5). hypothesis, post-conjugation curves overlapped 3A). experiments match 3B, 0.34 0.86 two- one-tailed t-tests, Thus, induce More generally, confirm (T0 Tpred), artifacts detected metric. 3. Individual two-tailed variable glucose (glu) casamino acid (caa) concentrations. Values % w/v. concentrations (C). average, error deviation, red 0.4% 0.04% w/v, respectively. Y-axis (Tpred/T0) acids. six Representatives replicates (see S10 day-to-day variability). R1, R1drd, pRK100 0.93, 0.79, 0.28, whereas RIP113, R6K, R6Kdrd 6.79e-05, 7.57e-05, 0.037, respectively, t-test, S3A). clinical p41, p168, p193, p283 37°C 7.10e-05, 2.10e-05, 0.021, 1.90e-05, n 4, 3, 2, cases except p283, averages scatter least replicates; p193 30°C 2.80e-04, 2.72e-04, scattered ratio Tpred/T0. linear regression best fit, R2 0.01 (shown left). deviation; type listed S3A. 3 [msb20209913-sup-0004-SDataFig3.zip] pR-specific differences arise demand, substrate consumed converted biomass) (Chudoba 1992) costs. Intuitively, inefficiently growing excess (Russell Cook, 1995; Russell, 2007) applied plasmid-related demands, resulting devote bulk biomass production (Low Chase, 1999), reallocating demands increase modulating well-established yields inefficient (Liu, Basan 2015), exogenous amino supplementation (Akashi Gojobori, 2002; Waschina 2016). Adapting approach leveraged trade-off 2019), (CAA) (0, 0.01, 0.1% w/v) (0.4% w/v). higher CAA increases rate, included fourth combination (0.0

Language: Английский

---

Recent advances in machine learning applications in metabolic engineering

Biotechnology Advances, Journal Year: 2022, Volume and Issue: 62, P. 108069 - 108069

Published: Nov. 25, 2022

Language: Английский

---

Antibiotic resistance in bacterial communities

Current Opinion in Microbiology, Journal Year: 2023, Volume and Issue: 74, P. 102306 - 102306

Published: April 11, 2023

Language: Английский

---

Modeling Microbial Community Networks: Methods and Tools for Studying Microbial Interactions

Microbial Ecology, Journal Year: 2024, Volume and Issue: 87(1)

Published: April 8, 2024

Abstract Microbial interactions function as a fundamental unit in complex ecosystems. By characterizing the type of interaction (positive, negative, neutral) occurring these dynamic systems, one can begin to unravel role played by microbial species. Towards this, various methods have been developed decipher communities. The current review focuses on qualitative and quantitative that currently exist study interactions. Qualitative such co-culturing experiments are visualized using microscopy-based techniques combined with data obtained from multi-omics technologies (metagenomics, metabolomics, metatranscriptomics). Quantitative include construction networks network inference, computational models, development synthetic consortia. These provide valuable clue roles interacting partners, well possible solutions overcome pathogenic microbes cause life-threatening infections susceptible hosts. Studying will further our understanding less-studied ecosystems enable design effective frameworks for treatment infectious diseases.

Language: Английский

---