Advances in application of machine learning to life cycle assessment: a literature review

Purpose

Life Cycle Assessment (LCA) is the process of systematically assessing impacts when there is an interaction between the environment and human activity. Machine learning (ML) with LCA methods can help contribute greatly to reducing impacts. The sheer number of input parameters and their uncertainties that contribute to the full life cycle make a broader application of ML complex and difficult to achieve. Hence a systems engineering approach should be taken to apply ML in isolation to aspects of the LCA. This study addresses the challenge of leveraging ML methods to deliver LCA solutions. The overarching hypothesis is that: LCA underpinned by ML methods and informed by dynamic data paves the way to more accurate LCA while supporting life cycle decision making.

Methods

In this study, previous research on ML for LCA were considered, and a literature review was undertaken.

Results

The results showed that ML can be a useful tool in certain aspects of the LCA. ML methods were shown to be applied efficiently in optimization scenarios in LCA. Finally, ML methods were integrated as part of existing inventory databases to streamline the LCA across many use cases.

Conclusions

The conclusions of this article summarise the characteristics of existing literature and provide suggestions for future work in limitations and gaps which were found in the literature.

     

Understanding the Physicochemical Characteristics and the Improved Enzymatic Saccharification of Corn Stover Pretreated with Aqueous and Gaseous Ammonia

Physicochemical characteristics of corn stover pretreated by soaking in aqueous ammonia (SAA) and low-moisture anhydrous ammonia (LMAA) were compared and investigated. The glucan digestibility of the treated biomass reached 90 % (SAA) and 84 % (LMAA). The LMAA pretreatment enhanced the digestibility by cleaving cross-linkages between cell wall components, whereas the SAA pretreatment additionally improved the digestibility by efficiently removing a major portion of the lignin under mild reaction conditions without significant loss of carbohydrates. Fourier transform infrared spectroscopy (FTIR), nuclear magnetic resonance (NMR), and gel permeation chromatography (GPC) revealed the structural and chemical transformations of lignin during the pretreatments. Both pretreatments effectively cleaved ferulate cell wall cross-linking that is associated with the recalcitrance of grass lignocellulosics toward enzymatic saccharification. Extracted lignin from SAA pretreatment was extensively depolymerized but retained “native” character, as evidenced by the retention of β-ether linkages.     

Exploring gene-environment interactions in Parkinson’s disease

The objective of this study was to explore combined effects of four candidate susceptibility genes and two exposures on Parkinson’s disease (PD) risk; namely, α-synuclein (SNCA) promoter polymorphism REP1, microtubule-associated protein tau (MAPT) H1/H2 haplotypes, apolipoprotein E (APOE) ε2/ε3/ε4 polymorphism, ubiquitin carboxy-terminal esterase L1 (UCHL1) S18Y variant, cigarette smoking and caffeinated coffee consumption. 932 PD patients and 664 control subjects from the NeuroGenetics Research Consortium, with complete data on all six factors, were studied. Uniform protocols were used for diagnosis, recruitment, data collection and genotyping. A logistic regression model which included gene-exposure interactions was applied. Likelihood ratio tests (LRTs) were used for significance testing and Bayesian inference was used to estimate odds ratios (ORs). MAPT (P = 0.007), SNCA REP1 (P = 0.012), smoking (P = 0.001), and coffee (P = 0.011) were associated with PD risk. Two novel interactions were detected: APOE with coffee (P = 0.005), and REP1 with smoking (P = 0.021). While the individual main effects were modest, each yielding OR < 1.6, the effects were cumulative, with some combinations reaching OR = 12.6 (95% CI: 5.9–26.8). This study provides evidence for the long-held notion that PD risk is modulated by cumulative and interactive effects of genes and exposures. Furthermore, the study demonstrates that while interaction studies are useful for exploring risk relationships that might otherwise go undetected, results should be interpreted with caution because of the inherent loss of power due to multiple testing. The novel findings of this study that warrant replication are the evidence for interaction of coffee with APOE, and of smoking with REP1 on PD risk. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Regenerating Eye Tissues to Preserve and Restore Vision

Download video (112MB)Help with mp4 files     

Optimization of nutrient parameters for lovastatin production by <Emphasis Type="Italic">Monascus purpureus</Emphasis> MTCC 369 under submerged fermentation using response surface methodology

Lovastatin, an inhibitor of HMG-CoA reductase, was produced by submerged fermentation using Monascus purpureus MTCC 369. Five nutritional parameters screened using Plackett–Burman experimental design were optimized by Box–Behnken factorial design of response surface methodology for lovastatin production in shake flask cultures. Maximum lovastatin production of 351 mg/l were predicted in medium containing 29.59 g/l dextrose, 3.86 g/l NH₄Cl, 1.73 g/l KH₂PO₄, 0.86 g/l MgSO₄·7H₂O, and 0.19 g/l MnSO₄·H₂O using response surface plots and point prediction tool of DESIGN EXPERT 7.0 (Statease, USA) software.     

A novel antioxidant and antiapoptotic role of omeprazole to block gastric ulcer through scavenging of hydroxyl radical

The mechanism of the antiulcer effect of omeprazole was studied placing emphasis on its role to block oxidative damage and apoptosis during ulceration. Dose-response studies on gastroprotection in stress and indomethacin-induced ulcer and inhibition of pylorus ligation-induced acid secretion indicate that omeprazole significantly blocks gastric lesions at lower dose (2.5 mg/kg) without inhibiting acid secretion, suggesting an independent mechanism for its antiulcer effect. Time course studies on gastroprotection and acid reduction also indicate that omeprazole almost completely blocks lesions at 1 h when acid inhibition is partial. The severity of lesions correlates well with the increased level of endogenous hydroxyl radical (*OH), which when scavenged by dimethyl sulfoxide causes around 90% reduction of the lesions, indicating that *OH plays a major role in gastric damage. Omeprazole blocks stress-induced increased generation of *OH and associated lipid peroxidation and protein oxidation, indicating that its antioxidant role plays a major part in preventing oxidative damage. Omeprazole also prevents stress-induced DNA fragmentation, suggesting its antiapoptotic role to block cell death during ulceration. The oxidative damage of DNA by *OH generated in vitro is also protected by omeprazole or its analogue, lansoprazole. Lansoprazole when incubated in a *OH-generating system scavenges *OH to produce four oxidation products of which the major one in mass spectroscopy shows a molecular ion peak at m/z 385, which is 16 mass units higher than that of lansoprazole (m/z 369). The product shows no additional aromatic proton signal for aromatic hydroxylation in (1)H NMR. The product absorbing at 278 nm shows no alkaline shift for phenols, thereby excluding the formation of hydroxylansoprazole. The product is assigned to lansoprazole sulfone formed by the addition of one oxygen atom at the sulfur center following attack by the *OH. Thus, omeprazole plays a significant role in gastroprotection by acting as a potent antioxidant and antiapoptotic molecule.     

Corrections to the Saffman-Delbruck mobility for membrane bound proteins

Recent experiments (Gambin, Y., R. Lopez-Esparza, M. Reffay, E. Sierecki, N. S. Gov, M. Genest, R. S. Hodes, and W. Urbach. 2006. Proc. Natl. Acad. Sci. USA. 103:2098-2102) have called into question the applicability of the Saffman-Delbrück diffusivity for proteins embedded in the lipid bilayers. We present a simple argument to account for this observation that should be generically valid for a large class of transmembrane and membrane bound proteins. Whenever the protein-lipid interactions locally deform the membrane, that deformation generates new hydrodynamic stresses on the protein-membrane complex leading to a suppression of its mobility. We show that this suppression depends on the protein size in a manner consistent with the work of Gambin et al.