To date Striatin family members

To date Striatin family members have not been associated with any catalytic function. They appear to act as scaffolds organizing different signalosomes that spatially coordinate multiple regulatory pathways. Striatin and SG2NA constitute a novel B-type subunit of Protein phosphatase 2A (PP2A), controlling the activity and specificity of the holoenzyme in different cellular compartments (Moreno et al., 2000uanduLechward et al., 2001). One multiprotein assembly named Striatin-interacting phosphatase and kinase (STRIPAK) that contains Germinal center kinase III (GCK III), Cerebral cavernous malformations 3 (CCM3) and PP2A has recently been described. It is involved in cell signaling, p-selectin control, vesicular trafficking, Golgi assembly, cell polarity, cell migration, neural and vascular development, cardiac function and apoptosis (Hwang and Pallas, 2014). STRIPAK complexes appear to have roles in certain diseases like diabetes, autism, cerebral cavernous malformation etc. (Hwang and Pallas, 2014). We have recently shown that SG2NA recruits Akt kinase and DJ-1, an antioxidant protein associated with Parkinson\’s disease and cancer, protecting cells from oxidative injury and promoting cell survival (Tanti and Goswami, 2014). Since SG2NA has several isoforms located in various cellular compartments, they are likely to have context specific versatile functions.
In the present study, we demonstrate that in addition to five spliced isoforms of SG2NA reported earlier, a sixth variant (52ukDa) is generated by the editing of 82ukDa transcript. Also, Sg2na transcripts are differentially polyadenylated and those with the longer UTR are found only in the brain. The expression level of SG2NAs varies among tissues and changes with tissue aging. Taken together, this study further highlights the diversity of regulation of expression of SG2NA, a scaffold protein with miscellaneous functions.
2. Materials and methods
Mouse neuroblastoma (Neuro2A) cell line were procured from National Centre for Cell Sciences, Pune, India (originally from ATCC, USA). Cells were cultured as a monolayer in DMEM containing 10% FBS, penicillin (90uU/ml), streptomycin (90uμg/ml) and amphotericin B (5uμg/ml) in a humidified, 5% CO2 containing incubator at 37uC.
The studies were performed with Swiss albino male mice. Tissue blots for 1 and 2umonth mice were done in three biological replicates while that for 9umonth was done for one mouse only. The use of animals was duly approved by the Animal Ethics Committee, Jawaharlal Nehru University.
2.3. Biochemical and molecular biology reagents
All biochemicals were procured from Sigma-Aldrich, USA unless mentioned otherwise. Reverse transcription kit was from Epicentre Biotechnologies, Madison, Wisconsin and Taq polymerase was from MBI Fermentas, Inc. All plastic wares were from Tarsons, Germany.
The 3 UTRs of Sg2na were cloned downstream to luciferase cDNA into pMIR-REPORT luciferase vector (Applied Biosystems, Inc., USA) at SacI HindIII sites.

Arabidopsis seeds over expressing Ph glucanase perform better

3.4. Arabidopsis seeds over-expressing Ph-glucanase perform better germination
Fig.6.uFunctional validation of Ph-glucanase in Arabidopsis. A) Diagrammatic representation of construct prepared in pCAMBIA1302 for ionophore inhibitor in Arabidopsis. B) Confirmation of Ph-glucanase expression in transgenic Arabidopsis plants. Total RNA was isolated from the leaves of the transgenic plants and first strand cDNA synthesis was done by reverse transcriptase with oligo dT primers. Specific cDNA of Ph-glucanase was amplified by using gene specific primers. 26SrRNA was used as a reference to show that equal amounts of RNA were used in the analysis. C) Germination of transgenic Arabidopsis seeds overexpressing Ph-glucanase at 20uu1uC. Errors bars represent uSE.Figure optionsDownload full-size imageDownload high-quality image (438 K)Download as PowerPoint slide
Fig.7.uPattern of germination of transgenic Arabidopsis seeds over-expressing Ph-glucanase as a function of temperature and ABA stress. A) Seeds were kept for germination at different temperatures (0, 5, 10, 15, 20 and 30uC) and B) in different ABA concentrations (0, 0.5, 1.0, 1.5, 2, 3, 5 and 10uμM) and the number of seeds germinated was recorded each day up to 8udays. Each value is a mean of three separate biological replicates. Error bars represent uSE. Seed germination percentages were analyzed using ANOVA to detect significant difference between means. Means were compared using Duncan\’s Multiple Range Test (DMRT) at Pu<u0.05.Figure optionsDownload full-size imageDownload high-quality image (229 K)Download as PowerPoint slide
Over-expression of Ph-glucanase gene from the high altitude medicinally important plant-Podophyllum, promoted Arabidopsis seed germination in different stress conditions like in low and high temperatures and ABA stress. Further biochemical and structural characterizations need to be done for better understanding and utilization of this important gene in agriculture as well as for commercial applications.
AcknowledgmentsThis work was supported by grants from the Department of Science and Technology (SR/FT/L-156/2004), New Delhi, India and the Council of Scientific and Industrial Research (CSIR), New Delhi, India in the form of Network Projects PlaGen (BSC-0107), SIMPLE (BSC-0109) and Developmental biology (MLP-072) at the CSIR-IHBT. The manuscript represents CSIR-IHBT communication number 3666. VD acknowledges CSIR, New Delhi for providing Senior Research Fellowship.
Appendix A.uSupplementary data
Supplementary tables.
uSupplementary Table 1: List of primers used for RACE reactions and Semi-quantitative RTPCR.Supplementary Table 2: Predicted secondary (using SOPMA) and tertiary structure (using SWISS-MODEL) of Ph-glucanase.Help with DOC filesOptionsDownload file ( K)
Supplementary material.Help with TXT filesOptionsDownload file (
SCIN, scinderin; MAP kinases P38, P38 mitogen-activated protein kinases; RNAi, RNA interference; cDNA, complementary DNA
Keywords

miRNA and human spermatogenic disorders There is increasing

2.4. miRNA and human spermatogenic disorders
There is increasing evidence to show a close relationship between altered or abnormal miRNA expressions in the testis and male reproductive disorders. This increases the significance of research relating to miRNAs. Microarray and real-time RT-PCR reveals that the expression of many miRNAs (miR-122, miR-34b, miR-34c-5p, miR-1973, miR-16, etc.) in the testis from asthenozoospermic or oligoasthenozoospermic patients differed to those in normozoospermic men (Abu-Halima et al., 2013). Similar research has shown that miR-122, miR-34c-5p, miR-146-5p, miR-374b, and other miRNAs were significantly decreased in azoospermia patients but increased in cases of asthenozoospermia (Wang et al., 2011). Comparative analysis of miRNAs expression patterns in the seminal plasma between patients with non-obstructive azoospermia (NOA) and healthy controls indicated that miR-141 (acting to depress CBL and TGFβ2), miR-7-1-3p (acting to depress RBL and PIK3R3) and miR-429 were all highly enhanced ( Wu et al., 2013). Another research group also found that patients with NOA showed a decrease in some miRNAs. These noted decreases included both the miR-17-92 and miR-371,2,3 clusters (Lian et al., 2009). Oligospermic patients have been found with the abnormal expression level of miR-100 and let-7b that possibly targets estrogen receptor-alpha (ERα) (Abhari et al., 2014). MiR-383 is down-regulated via a feedback interaction with the fragile X mental retardation protein (FMRP) in the testes of infertile patients who had suffered from sperm maturation arrest (Tian et al., 2013). Some of those specific miRNAs are probably conserved across species as part of normal spermatogenic function and thus deserve further study. In addition, even after vasectomy, the miRNA profile in the epididymis and seminal microvesicles will undergo ether partially reversible or non-reversible changes (Belleannee et al., 2013b).
3. piRNA and spermatogenesis
3.1. Discovery and characteristics of piRNAs
Mammalian piRNAs are predominantly testis-specific, though piRNAs generated from other animals (e.g. fly, nematode and zebrafish) also can be found in the ovaries (Houwing et al., 2007uanduZhou et al., 2010). However, even in organisms expressing piRNA in the germ Hexa His tag of both genders, the testis-derived piRNAs and the ovary-derived piRNAs appear to have evolved independently (Billi et al., 2013). Additionally, allele-specific DNA methylation pattern differs in human PIWI gene loci, PIWIL1 and PIWIL2, between normal and infertile males ( Friemel et al., 2014).
In the following sections, we will review the known characteristics and possible functions of piRNA and PIWI proteins, especially relating to their potential roles in spermatogenesis.
3.2. PIWI proteins and their roles in spermatogenesis
The biogenesis and function of piRNA is closely linked to that of the PIWI protein subfamily. Surprisingly, the participation of PIWI Proteins had been revealed in spermatogenesis at a much earlier time than the later identification of piRNA itself. In 1997, the mutation of a Drosophila PIWI homolog gene, piwi, was found to lead to a suppression of germ stem cell differentiation ( Lin and Spradling, 1997). In a manner similar to piRNAs, PIWI proteins are expressed during the development of both egg cells and sperm cells in the gametogenesis of Drosophila ( Lin and Spradling, 1997). In a contrasting focus to the studies of PIWI proteins in mammals, which tend to focus on spermatogenesis, much of the research relating to Drosophila piRNAs have their focus upon the ovaries. Another two fly PIWI homologs, Aub and Ago3 are also involved in piRNA generation. This will be discussed in the next section. Piwi seems to have a similar but less critical role compared to Aub in piRNA biogenesis. Its localization requires the assistance of Aub and/or secondary piRNA ( Brennecke et al., 2007). Surprisingly, piwi knockdown will decrease the levels of both Aub- and Ago3-assocaited RNAs in Drosophila ovaries ( Rozhkov et al., 2013). Piwi is also thought to have an additional post-transcriptional regulation ability in germ cells (Rozhkov et al., 2013). Relating to Drosophila oogenesis, in certain somatic cells Piwi even appears to regulate germ stem cell division ( Cox et al., 1998). Such properties of PIWI proteins are also conserved in C. elegans ( Cox et al., 1998). Our lab identified three piwi-like genes in the testis of Portunus trituberculatus, with different temporal and spatial expression patterns ( Xiang et al., 2014). The zebrafish homolog Ziwi appears to exert a similar role and its mutations lead to germ cell apoptosis (Houwing et al.,, 2007). The Xenopus homolog Xiwi was also found in oocytes and may have an additional role in the sorting of piRNAs ( Lau et al., 2009). The various identified PIWI homologs of different organisms have been more fully summarized in Thomson and Lin;s review paper (Thomson and Lin, 2009).

AbyssBox Biological interactions Deep sea

AbyssBox; Biological interactions; Deep-sea observatory; Eiffel Tower edifice; Feeding behaviour; Experimental research; Video imagery; Time series; Lucky Strike; Mid-Atlantic Ridge; 37°17″N; 32°16.3″W
1. Introduction
Hydrothermal vents are located along mid-ocean ridges, back-arc basins and volcanic seamounts where seawater percolates through the thin oceanic crust and is ejected as hot fluids with high concentrations of reduced sulphur, methane and metals (e.g., iron, copper and zinc). They host highly productive communities fuelled primarily by chemosynthetic microbial production dependent on ephemeral fluxes of sulphide- and methane-rich emissions ( Tunnicliffe, 1991). The mixing of these fluids with seawater causes steep environmental gradients resulting in small-scale temporal (e.g., minute) and spatial (e.g., cm) physico-chemical variations ( Johnson et al., 1986), which can create a large number of microenvironments on a single edifice ( Sarradin et al., 1998, Sarrazin et al., 1999 and Luther et al., 2012). To date most of the knowledge about the functioning of these remote ecosystems results from, at best, yearly sea-going cruises. Consequently, little is known about the temporal and small spatial variations in third structure and how organisms respond to changes in local environmental conditions ( Cuvelier et al., 2012, Cuvelier et al., 2014 and Sarrazin et al., 2014).
Most ecological studies emphasise the importance of the high spatial variability of abiotic factors in terms of fluid flow, temperature and chemical composition, and the physical structure of the mineral substrate in controlling benthic species distribution within a single vent structure (Hessler et al., 1988, Tunnicliffe, 1991, Sarrazin et al., 1997, Sarrazin et al., 2002, Shank et al., 1998, Cuvelier et al., 2012 and Cuvelier et al., 2009). Nevertheless, the mechanisms by which environmental conditions shape vent communities appear quite complex. According to the correspondence between physico-chemical gradients and faunal zonation, physiological tolerance and nutritional requirements were suggested to be the main direct pathways of environmental control (Desbruyères et al., 1998, Fisher, 1998, Lee, 2003 and Bates et al., 2005). Environmental conditions can also indirectly influence species distribution by controlling biological interactions within and among species, including predation (Micheli et al., 2002 and Sancho et al., 2005), competition and species behaviour (Bates et al., 2013). While biotic interactions can act independently of physico-chemical factors within a habitat type (Mullineaux et al., 2000, Govenar et al., 2005 and Lenihan et al., 2008), those processes seem to vary along a gradient of flow intensity with facilitation processes (e.g., providing refuge for new recruits) occurring at the periphery of vents and inhibition processes (e.g., grazing of new recruits, competition for space) dominating high diffuse flow areas ( Mullineaux et al., 2003). Finally, variations in fluid flow can significantly control population dynamics including growth (Schöne and Giere, 2005) and reproduction ( Copley et al., 2003, Kelly and Metaxas, 2007, Sheader and Van Dover, 2007, Nye et al., 2013 and Sheader and Van Dover, 2007), but studies are scarce. Indeed, the lack of direct observations and long time-series (Glover et al., 2010) combined with the difficulty to maintain vent animals in aquaria led to a major gap in our current knowledge of species auto-ecology (or species ecology; i.e., relationship between a single species and its environment).
In situ imagery is a good means to investigate species spatial distribution because it provides access to living species in their natural habitats and has been widely used since the discovery of hydrothermal vent communities ( Hessler et al., 1985, Tunnicliffe and Juniper, 1990, Grehan and Juniper, 1996, Copley et al., 1997, Copley et al., 2007, Sarrazin et al., 1997, Shank et al., 1998, Cuvelier et al., 2009, Cuvelier et al., 2011a, Cuvelier et al., 2011b and Cuvelier et al., 2012). In addition, imagery methods mitigate environmental impact and disturbance related to sampling (Tunnicliffe, 1990). With the development of deep-sea observatories, we can now describe patterns and the underlying processes at sub-annual scales using video cameras ( Cuvelier et al., 2012, Cuvelier et al., 2014 and Sarrazin et al., 2014). They provide high-resolution video time-series that help to document small-scale assemblage dynamics and to study organisms? growth, faunal succession, biological interactions and species/communities? responses to environmental changes ( Juniper et al., 2007 and Sarrazin et al., 2007). Continuous video data have already been successfully used to assess species behaviour ( Chevaldonné and Jollivet, 1993, Bates et al., 2005, Grelon et al., 2006, Robert et al., 2012 and Tunnicliffe et al., 2013) but those studies are restricted to polychaetes and gastropods and are limited in time. In this context, an ecological module called TEMPO was developed to specifically study long-term deep-sea communities? dynamics at hydrothermal vents ( Sarrazin et al., 2007 and Auffret et al., 2009) and was deployed in 2010 on the Mid-Atlantic Ridge within the Lucky Strike vent field ( Cannat et al., 2011 and Cola?o et al., 2011).

Material and methods Sequence analysis

2. Material and methods
2.1. Sequence analysis of LasR
The Amino i thought about this sequence of LasR protein of P. aeruginosa strain PAO1 was obtained from NCBI (Sequence ID NP_250121.1) ( http://www-ncbi-nlm-nih-gov.ezproxy.lib.ncyu.edu.tw/protein/15596627, n.d.). The crystal structure of the N-terminal autoinducer binding domain of LasR protein (amino acid residues 1 to 169) from P. aeruginosa was available in the Brookhaven Protein Data Bank (PDB) (PDB ID: 3IX3, chain — A) (http://www-ncbi-nlm-nih-gov.ezproxy.lib.ncyu.edu.tw/protein/15596627, n.d.). However, in order to find the mode of interactions between the LasR protein and the DNA, the structure of the entire LasR protein was needed. Therefore, the amino acid sequence of the rest of the part of the LasR protein was used to search PDB to find suitable template(s) for homology modeling using the tool BLAST (Berman et al., 2000). From this BLAST search, the best match was found to be the crystal structure of QscR, bound to N-3-OxoDodecanoyl-L-Homoserine Lactone, (PDB code 3SZT, chain — A) (Altschul et al., 1990) with 30% sequence identity, 97% query coverage and E-value of 1uu10a25.
However, the other similar structures which were obtained from BLAST search results were as follows:1.1FSE, chain — A with 37% sequence identity, 28% query coverage and E-value of 2uu10u4. It is the crystal structure of Bascillus subtilis regulatory protein gene.2.4Y15, chain — A with 30% sequence identity, 75% query coverage and E-value of 3uu10u16. It is the crystal structure of Sdia in complex with 3-oxo-c6-homoserine Lactone.3.3ULQ, chain — B with 32% sequence identity, 25% query coverage and E-value of 0.09. It is the crystal structure of the Anti Activator Rapf complexed with the response regulator coma DNA binding domain.
2.2. Modeling of LasR monomer
The protein LasR is a 239 residue long protein. The homology modeling of LasR protein was done using the HHPred web server (Söding et al., 2005). The HHPred web server uses a multi-template modeling approach and the templates, used for the homology modeling of LasR, were 3SZT_A, 3IX3_A, 1FSE_A, 4Y15_A and 3ULQ_B. These structures were also identified by BLAST to be the suitable templates for homology modeling of LasR protein. The final model of LasR protein spanning the amino acid residues 170 to 239 was built using the aforementioned templates piece by piece. The built model of the LasR protein spanning the amino acid residues 170 to 239 was joined to the existing X-ray crystal structure of LasR protein (3IX3, chain — A) which had amino acid residues 1 to 169. This was done in Discovery i thought about this Studio 2.5 (DS). The root mean squared deviation (RMSD) of the built model was checked by superposing the coordinates of the modeled protein on to the template 3SZT, chain — A and it was found to be 0.7?. The stereo-chemical fitness of the modeled LasR protein was checked by PROCHECK (Laskowski et al., 1993) and Verify3D (Eisenberg et al., 1997) and Ramachandran Plots were drawn. The structure was found to be stereo-chemically fit. More than 99.58% of the amino acid residues in the final modeled structure had a 3Da2D score >u0.2 as specified by Verify3D for a good computational model (Eisenberg et al., 1997). No amino acid residues were found to be in the disallowed regions of the Ramachandran Plot (Supplementary Fig. 1). The PROSA value of this model was u6.63 (Supplementary Fig. 1) which represents a good model quality for homology models (Wiederstein and Sippl, 2007).

Research on task scheduling problems Task

2.2. Research on task scheduling problems
Task scheduling can be considered a subproblem of workforce planning, see the framework for workforce planning in [17]. They separate the task scheduling problem from the superordinate planning problems by combining “individual tasks into task sequences that could usefully be carried out by one person” to derive an aggregated demand. There are dedicated literature reviews on workforce planning in health care [18] and [19], but they leave a message focus leave a message solely on staffing decisions. As they consider an aggregated demand for workers rather than individual tasks, they do not consider detailed scheduling with respect to the clients’; time preferences.
De Bruecker et al. [8] provide a broad overview of papers on workforce planning, including both, planning of individual tasks and planning with hierarchical skills, but none of the references therein considers the same combination of assumptions as the problem at hand. Baker and Scudder [20] review scheduling problems with homogeneous resources, earliness/tardiness penalties, and a preferred starting time but without a hard time window. Task scheduling problems with multiple resources which take either different qualifications (also referred to as skills) or time preferences into account can be found in the literature for different areas of application:
Task scheduling with different qualifications
Bellenguez-Morineau and Néron [21] present a project scheduling problem that assumes a given workforce with hierarchical skill levels. As they do not consider time preferences for individual tasks, the objective is to minimize the makespan. Krishnamoorthy et al. [22] present a task scheduling problem with hierarchical skill levels. They assume fixed start times for all tasks as a hard constraint. Their objective is to find the minimum required workforce to obtain a feasible schedule. This paper, on the contrary, assumes a given workforce composition and variable start times for tasks.
Schimmelpfeng et al. [23] present a task scheduling approach for rehabilitation hospitals with different qualifications and precedence constraints between tasks, but they do not consider time preferences for individual tasks. A general difference between a hospital and a nursing home is that patients usually visit hospitals only for a short period of time, therefore hospitals can focus on high resource utilization rather than on meeting time preferences.
Task scheduling with time windows
Gertsbakh and Stern [24] discuss task scheduling with time windows for a homogeneous workforce. They do not consider earliness/tardiness penalties. The objective is to find the minimum required workforce to obtain a feasible schedule.
Mankowska et al. [25] discuss the Home Health Care Routing and Scheduling Problem as a Vehicle Routing Problem with Time Windows. Clients are visited in their homes, therefore sequence-dependent travel times are taken into account. They consider time preferences and different skills. However, they assume half-open time windows that do not allow earliness while tardiness is penalized, but not limited. The objective is to minimize the weighted sum of total travel times, total tardiness, and maximum tardiness.

Model formulation In our model

2.1. Model formulation
In our model, we are given a set of demand points NN and a set of possible done locations MM. For each demand point jj, we have a given demand djdj. This djdj should be a measure for the number of calls within demand point jj. See, for example, Channouf et al. [23], and Setzler et al. [24] for EMS call volume forecasting methods. Each base location has a capacity bibi, which is the maximum number of ambulances that may be located at that station. In total we are allowed to use at most ββ base locations. The total number of available ambulances is bb. The busy fraction of an ambulance is denoted by qq. For each combination of a demand point jj and a base location ii, we have a probability wijwij that an ambulance departing from base ii will reach demand point jj within the time threshold. For fixed demand point jj, given wijwij, we can order the base locations from the closest to the furthest for this demand point. Let aijaij denote the index of the base location that is in position ii in this ordering for demand point jj. Similarly, let ranking(i,j) be the ranking of base ii in the ordering of demand point jj. So, by definition we have ranking(aij,j)=i.
The most straightforward way of modeling our problem is to introduce a decision variable xixi denoting the number of ambulances assigned to location ii. The expected coverage of demand point jj in terms of xixi is then equation(1)cj(x)=∑i∈Mq∑k<ranking(i,j)xakj(1?qxaij)waijj. Clearly, this formulation is not linear in the decision variables. When solving larger instances, this can result in longer computation times. To avoid this, we present a different formulation for which the objective is linear in the decision variables.
In order to formulate a linear model, we introduce a new binary decision variable zijkzijk indicating whether the kkth preferred, with respect to wijwij, ambulance for demand point jj is located at base location ii. If, for example, base location 1 is the closest one for demand point 2 and we have three ambulances located at that base location, we get z121=z122=z123=1z121=z122=z123=1. Additionally, we introduce a binary variable yiyi, which has value 1 if and only if at least one ambulance is located at base location ii. This variable is needed to limit the number of base locations that is used.
Using these decision variables we are able to formulate our model as follows: equation(2)maxc(z)=∑j∈Ndjcj(z)withwithequation(3)∑k=1bzijk≤xi?i∈M,j∈N,equation(4)∑i∈Mzijk=1?j∈N,k≤b,equation(5)∑i∈Myi≤β,equation(6)xi≤biyi?i∈M,equation(7)∑i∈Mxi=b,equation(8)yi,zijk∈ 0,1 ?i∈M,j∈N,k≤b,equation(9)xi∈N?i∈Mandandcj(z)=∑k=1b(1?q)qk?1∑i∈Mzijkwij?j∈N. The objective is to maximize the expected coverage over all demand points. This is defined as the sum of the coverages that can be provided to an individual node by the whole system, cj(z)cj(z), multiplied by the total demand generated at this node, djdj. The value cj(z)cj(z) is calculated by conditioning on the number of unavailable ambulances. The probability that the kkth ambulance is the first available one equals (1?q)qk?1(1?q)qk?1. If the kkth preferred ambulance is located at location ii, we obtain an expected coverage of wijwij. Constraints (3) state that no more than xixi ambulances may be assigned to base ii. This makes sure that the zijkzijk’;s have the right value. Constraints (4) ensure that the kkth preferred ambulance of demand point jj is located at no more than one base location. In order to design a realistic system, we add a limitation on the maximum number of base locations by constraint (5). This constraint is not included in Ingolfsson et al. [18]. They assume that the set of bases is fixed. Constraints (6) guarantee that the number of vehicles located at each station does not done exceed its capacity. Finally, constraint (7) states that no more than bb ambulances are used.

Fig Estimated events on gate no Figure optionsDownload full size

Fig. 2. Estimated events on gate no. 4.Figure optionsDownload full-size imageDownload as PowerPoint slide
As it description has been illustrated in the previous section, there is a high impact of ETA and EOBT disturbances on a low gate occupancy factor when the original planning strategy tries to maximize the gate occupancy factor. The use of remote points as a solution to non-anticipated disturbances impacts drastically on handling resources availability and handling operational costs (and sometimes turnaround times) since these last-moment remote gate reassignments (after the taxiing) generate an extra workload to the handling company (passenger buses and air stairs together with other resources required). Furthermore, capacities of ramp facilities can act as a hard constraint to other handling operators servicing aircraft which generates a domino effect on new departure delays.
3. Infrastructure overdimension for disturbance mitigation: Emergent dynamics
To minimize last minute gate reassignments to remote points, some airports have increased the number of terminals and contact points to tackle poor arrival predictability. Unfortunately, an increased amount of gates leads to a considerable economic investment (non-added value operation), but also leads to undesirable emergent dynamics caused mainly by longer distances between terminals. Fig. 6 shows (RP is used for Remote Point or off-pier stands, and CP for Contact Points or stands at the terminal) an airport layout, an X-shaped pier configuration [20], [21] and [22], in which it is easy to check that a greater amount of gates implies a greater distance between gates, and lower productivity of handling resources due to displacements.
Fig. 6. An airport layout: X-shaped pier configuration.Figure optionsDownload full-size imageDownload as PowerPoint slide
Airport infrastructure oversizing is often a consequence of underestimating the influence of operational issues during the strategic design of the airport. The gate oversizing approach usually deals with hard to solve drawbacks in both areas:?Strategic:-The increment of contact points usually leads to more complex airport layouts.-Due to an unnecessary increment of gates, employees and passengers must cover longer distances.?Operational:-Airlines constraints: some airlines prefer to allocate their flights in the same area to reduce passenger handling costs.-Passenger constraints: transfer connectivity is highly dependent on the distances between terminals when flights are late on the arrival.-Ramp capacity constraints: most handling companies prefer to concentrate all the operations to be served in the same time horizon in a narrow area for the different flights in order to avoid long distance equipment movements in the platform.-ATC constraints: complex layouts increase the number of taxiways to be properly managed to support the safe movement of the aircraft.
Due to the influence of strategic design decisions on the performance of most airport operations (handling, taxiing, transfers), tube-within-a-tube system is important to understand the impact of perturbations on the throughput of operations.

Data and aggregation The CGE MIRAGE model is a flexible

3.2. Data and aggregation
The CGE MIRAGE model is a flexible tool that can be tailored to respond to different policy questions. In the present case, we model the agri-food sector in as much detail as possible. We consider 31 different sectors (of which 16 are related to agri-food products), and 24 geographical areas. Detailed aggregations of sectors are provided in Table 4 and of regions in Table 5.
Table 4.
Sector aggregation.Agri-foodIndustryCerealsClothingVegetable, fruits and nutsChemical, rubber and plastic productsOil seedsMetal productsSugar, sugar cane and beetTransport equipmentPlant-based fibresElectronic devicesOther cropsMachinery and EquipmentLive animalsOther manufacturingRed meatServicesWhite meatBusiness ServicesMilk and dairy productsTransportOther animal productsFinance and insuranceForestryRecreation and other servicesFishingPublic administrationVegetable ppar inhibitor and fatsOther servicesBeverages and tobacco productsEnergy and other primaryOther food productsEnergyOther primaryFull-size tableTable optionsView in workspaceDownload as CSV
Table 5.
Country aggregation.European Union (28)Other countriesUSBrazilOther NAFTAArgentinaCanadaEuropean Free Trade AreaMexicoOther EuropeTPP MembersRussian FederationAustralia and New ZealandTurkeyChile and PeruOther Middle EastSingapore, Malaysia and VietnamNorth AfricabJapanOther AfricaOther ASEANaRest of the WorldcTPP Potential membersChinaIndiaKoreaOther AsiaOther Latin AmericaNote:a: Cambodia, Indonesia, Lao PDR, Philippines, Thailand, Rest of Southeast Asia;b: Egypt, Morocco, Tunisia, Rest of North Africa;c: Rest of Oceania, Rest of North America, Rest of the World.Full-size tableTable optionsView in workspaceDownload as CSV
MIRAGE relies on the Global Trade Analysis Project (GTAP) database for social accounting matrices (version 8.1). Tariffs are taken from the MAcMap-HS6 database (Guimbard et al., 2012) for years 2007 and 2010, and AVEs for NTMs in goods are our own estimations (see above). AVEs for NTMs in the service sector are taken from Fontagné et al. (2011).
Although we have information on tariffs and NTMs at the HS-6 level (respectively τr,s,i and αs,i) the CGE MIRAGE model is not defined at this level of detail since proteins would require very complex data. Instead, the model only includes 24 regions and 31 sectors. When aggregating tariffs and AVEs at the country-HS6 level (r, s, i) to the MIRAGE aggregation level (R, S, I), we take advantage of two valuable pieces of information. First regarding NTMs, we consider the contribution of an HS6 line to the aggregate AVE only if at least one NTM is actually implemented in the (r, s, i) trade flow, using the dummy variables δr,s,i constructed from the WTO notifications. Second, the MAcMap-HS6 database provides information on the weights ωr,s,i of each disaggregated flow which can be used to further aggregate the tariffs and NTMs from the HS6 to the MIRAGE level. These weights are computed using the reference group method which requires each country to be allocated to one of five world regions (the reference groups) with similar characteristics, using hierarchical clustering analysis. The weight of each flow is the share of good i in the imports of the whole reference group originating from region r, scaled by the size of the imports of country s in its reference group (for more details, see Guimbard et al., 2012). The advantage gained by deriving these weights compared to simply weighting by trade flows is that they take account of at least part of the prohibitiveness of certain transaction costs. Letting G(s) denote the reference group of country s, and (ρ, ι) denote the country-HS6 level:equation(3)ωr,s,i=Mr,Gs,i∑ρι∈riMρ,s,ι∑ρι∈riMρ,Gs,ι?∑ι∈iMs,Gsιequation(4)aveR,S,I=∑rsi∈RSIωr,s,iδr,s,iαs,i∑rsi∈RSIωr,s,iequation(5)tariffR,S,I=∑rsi∈RSIωr,s,iτr,s,i∑rsi∈RSIωr,s,i

In recent years artificial intelligence models and machine learning such

In recent years, artificial intelligence models and machine learning such as support vector machine (SVM), adaptive neuro-fuzzy inference system (ANFIS) and artificial neural network (ANN) have been gaining popularity because of their high flexibility and proven prediction abilities (Abbasi et al., 2013, Abbasi et al., 2014 and Antanasijevi? et al., 2013). Intelligent models are shown to be capable of predicting MSW generation on long, medium and short term scales (Abbasi et al., 2014, Abdoli et al., 2012 and Jalili Ghazi Zade and Noori, 2008). However, there is limited information about monthly prediction of MSW generation as well as optimal algorithm for this purpose. This paper will review the state of the art of the intelligent modelling approaches for MSW generation forecast and then apply them to a real case scenario in order to identify the most suitable algorithm to predict MSW collected by kerbside service on a medium-term scale.
2. Application of artificial intelligence in forecasting MSW generation
Advanced artificial intelligence forecast systems have shown superiority to conventional models in engineering problems as well as in waste management research (Abdoli et al., 2012). Recent research in this topic focused on using artificial intelligence models to deal with the non-linearity of the historical data. In this section, application of the techniques including ANN, ANFIS, SVM and k-nearest neighbours (kNN) will be reviewed in the field of MSW generation.
2.1. Artificial neural network
Artificial neural networks are cellular information processing systems designed and developed on the basis of the perceived notion of the human smoothened signaling pathway and its neural system (Firat et al., 2010). One of the most beneficial and significant features of ANN in forecasting is its learning ability. ANN can construct a complex nonlinear system through a set of input/output examples. Consequently, ANN has been successfully employed in nonlinear system modelling (Firat et al., 2010). Accordingly, the nonlinear structure of MSW generation makes the ANN an ideal candidate for forecasting waste generation. Literature survey returned applications of ANN to forecast MSW generation in short, medium and long-term periods (Abbasi et al., 2013, Abdoli et al., 2012 and Noori et al., 2010).
Ordóñez-Ponce et al. (2006) employed multi-layer perceptron neural network to predict long-term generation rate of MSW in Chile. Using a range of variables which covered socio-demographic, economic, geographic and waste-related factors, ANN was able to predict waste generation with great accuracy (R2 = 0.819). Ordóñez-Ponce et al. (2006) concluded that population, percentage of urban population, years of education, number of libraries and number of indigents were the most important factors which affected waste generation in Chile.
The ability of ANN to predict short-term MSW generation was also examined by other researchers (Noori et al., 2009a and Noori et al., 2009b). These studies focused on forecasting MSW generation by analysing time series of waste generation rather than analysing effective factors in waste generation. Results showed that feed-forward ANN with one hidden layer and 16 neurons was the best structure to forecast short-term waste generation rates (Noori et al., 2009a and Noori et al., 2009b). However, ANN accuracy may suffer when faced with large database due to the effect of irrelevant, redundant and noise in the data. Therefore, different input selection methods such as smoothened signaling pathway principal component analysis, wavelet transform and gamma test were introduced to deal with accuracy loss (Noori et al., 2009a and Noori et al., 2009c). Although ANN model has good ability to forecast MSW generation, its performance suffers because of its tendency to over-fitting training, local minimum, and poor generalization.
2.2. Adaptive neuro-fuzzy inference systems
Adaptive neuro-fuzzy inference systems (ANFIS) are a well-known data driven modelling technique that combines ANN and fuzzy logic. ANFIS is composed of two parts, antecedent and conclusion, which are connected to each other by fuzzy rules based on the network form. Limited attempts, 3 studies, to predict waste generation using ANFIS were found in the literature. The studies compared the performance of ANN and ANFIS models ability to predict MSW generation (Chen and Chang, 2000, Noori et al., 2009c and Tiwari et al., 2012). Tiwari et al. (2012) suggested that ANFIS is a more reliable model than ANN for forecasting the aggregate impact of economic trend, population changes, and recycling on solid waste generation. Chen and Chang (2000), on the other hand, demonstrated the ability of ANFIS to forecast waste generation with limited input data. Chen and Chang (2000) and later Noori et al. (2009c) applied fuzzy goal regression method to improve the overall prediction accuracy of ANFIS.