It is interesting to note that the increase in water discharge tr

It is interesting to note that the increase in water discharge transiting the interior of the delta have combined with the decrease in sediment load due to damming to keep sediment load directed toward the delta plain quite constant with ∼2.1 MT/yr for the Danube natural system

load at the delta of ∼70 MT/yr and ∼2.5 MT/yr for the anthropogenic system when the load decreased to ∼25 MT/yr. These numbers highlight the fact that due to the increase in density of human-dug canals sediment trapping on the delta plain Crenolanib supplier has become a significant part of the present sediment budget of the delta (i.e., >10%). In the same time, these numbers suggest that the main impact of Crizotinib the increasing fluvial sediment deficit would be expected at the coast. If we assume that sediments that enter the interior of the delta from the main distributaries, either as overbank flows or via the narrow and shallow secondary canal network, do not escape back into the main distributaries, the sediment trapped in the interior of the delta can be estimated. This tenet is a reasonable one if we take into account almost all branches of the canal network end in or cross lakes that act as sediment traps. Making the supplementary

assumption that most, if not all, of this sediment feeds the internal fluvial delta rather than the marine delta, because canal Y-27632 2HCl density is insignificant in the latter, we estimate the average sediment flux changed from 0.07 in natural conditions to 0.09–0.12 g/cm2 today when distributed uniformly across for an area the entire internal delta plain (∼2800 km2

or ∼2000 km2 without polders). The figures would be somewhat smaller when consider the losses to areas of the marine delta plain that do have some canals. However, these numbers ignore organic sedimentation that is expected to be significant in the internal delta. The flux estimates above translate into sedimentation rates of 0.5–0.8 mm/yr if we use a dry density of 1.5 g/cm3 for water saturated mixed sand and mud with 40% porosity (Giosan et al., 2012). In natural conditions, most of the internal delta plain was submerged with the exception of the levees of major and minor distributaries suggesting a sediment starved environment (Antipa, 1915). In anthropogenic conditions, the situation is probably similar with sediments rather than being spread evenly across the delta, accumulating close to the secondary channel network or in lakes fed by this network.

, 2012) Here we present three typical case studies where the lac

, 2012). Here we present three typical case studies where the lack of terrace maintenance characterizing the last few years has increased the landslide risk. The case studies are located in three different Italian regions (Fig. 5): Cinque Terre (a), Chianti Classico (b), and the Amalfi Coast (c). The Cinque Terre (The Five Lands)

is a coastal region of Liguria BGB324 clinical trial (northwestern Italy), which encompasses five small towns connected by a coastal pathway that represents an important national tourist attraction. Since 1997, this rocky coast with terraced vineyards has been included in the “World Heritage List” of UNESCO for its high scenic and cultural value. More recently, in 1999, it has become a National Park for its environmental and naturalistic relevance. Due to the morphological characteristic of this area, the landscape is characterized by terraces, supported by dry-stone walls, for the cultivation of vineyards. These terraces are not only an important cultural heritage but also a complex system

of landscape engineering (Canuti et al., 2004). However, the recent abandonment of farming and the neglect of terraced Selleckchem GDC0199 structures have led to a rapid increase in land degradation problems, with serious threats to human settlements located along the coast, because of the vicinity of mountain territories to the coastline (Conti and Fagarazzi, 2004). The instability of the dry-stone walls and the clogging of drainage channels are now the main causes behind the most frequent landslide mechanisms within the Cinque Terre (rock falls and topples along the sea cliffs and earth slides and debris flows in the terraced area) (Canuti et al., 2004). Fig. 6 shows the typical terraced landscape of the Cinque Terre subjected during to extensive land degradation: the dry-stone walls abandoned or no longer maintained have collapsed due to earth pressure or shallow landslides. The landslide processes and related terrace failures illustrated in Fig. 6 were triggered by an intense rainfall event that occurred on 25 October

2011, where more than 500 mm of cumulated rainfall was observed in 6 h. Another example of the acceleration of natural slope processes caused by anthropogenic activity is represented by the Chianti hills in Tuscany (Canuti et al., 2004). The terraced area of Tuscany is particularly vulnerable to the combination of geological and climatological attributes and economic factors associated with specialized vineyards and olive groves. The farming changes that have taken place since the 1960s through the introduction of agricultural mechanization, the extensive slope levelling for new vineyards and the abandonment of past drainage systems, have altered the fragile slope stability, generating accelerated erosion and landslides, particularly superficial earth flows and complex landslides (Canuti et al., 2004). Different authors (Canuti et al., 1979, Canuti et al., 1986 and Canuti et al.

4 and 100%, for the model based on normalized spectra, and 94 6 a

4 and 100%, for the model based on normalized spectra, and 94.6 and 95%, for the model based on first derivatives. Such results confirm that DRIFTS provides satisfactory discrimination between defective and non-defective roasted coffees, demonstrating its potential for detection of defective beans in mixtures with non-defective ones after roasting. Regarding the application of such methodology for routine analyses

of roasted coffee quality, further studies are still necessary, involving a trained panel of coffee tasters, to establish the minimum amount, if any, in which defective beans can be introduced to a non-defective coffee batch and changes in the beverage quality would still not be perceived in relation to one without defective beans. With the minimum amounts effectively established, mixtures of defective and non-defective roasted beans can be suitably prepared and duly tested for the find more discrimination capability of the developed models. The feasibility C59 wnt of employing DRIFTS as a methodology for discrimination between defective and

non-defective roasted coffees was evaluated. The obtained spectra were similar, with small differences in absorbance intensity between non-defective and defective coffees. PCA results based on DR spectra and first derivatives indicated separation of the samples into four major groups: non-defective, black, dark sour and light sour, with immature beans scattered among the sour samples. LDA classification models, based on absorbance readings and derivatives at eight wavenumbers (2924, 2852, 1743, 1541, 1377, 1076, 910 and 816 cm−1), provided separation of the samples into five groups: non-defective, black, dark sour, light sour and immature beans. Average recognition

and prediction abilities ranged from 79 to 96% and from 80 to 100%, respectively. Discrimination functions for generic classes of defective and non-defective coffee samples were also developed. For these generic models, recognition and prediction abilities ranged from 95 to 97% and from 95 to 100%, respectively. The results obtained in the present study confirm that DRIFTS provides satisfactory levels of discrimination between defective and non-defective coffee beans after roasting. The authors acknowledge financial support from the following Brazilian Government Agencies: CNPq from and FAPEMIG. “
“Events Date and Venue Details from COFE 2012 – 11th Conference of Food Engineering 2–4 April 2012 Leesburg, Virginia USA Email:[email protected] Food Colloids 2012 15–18 April 2012 Copenhagen, Denmark E-mail: Richard Ipsen: [email protected] NEFood: 1st North European Congress on Food 22–24 April 2012 St. Petersburg, Russia Internet:http://nefood.info 8th International Conference on Diet and Activity Methods 8–10 May 2012 Rome, Italy Internet:http://www.icdam.org 11th International Hydrocolloids Conference 14–17 May 2012 Purdue University, USA Internet:http://www.international-hydrocolloids-conference.

Nearly all melanoma cell lines tested to date have shown pRb path

Nearly all melanoma cell lines tested to date have shown pRb pathway alterations due to p16 or pRb deficiency, cdk4 mutation, or cyclin D1 overexpression [18] and [29]. In all types of melanoma, the most frequently amplified region is chromosome 11q13 [2] and [3], which harbors the cyclin D1 gene. Although cyclin D1 is a well-known growth promoter, it may also function as a survival factor for tumor cells [27] and [31]. Cyclin D1 amplification or overexpression is a crucial event that leads to melanoma progression

[10] and is associated with high proliferation rates in these tumors [17] and [29]. Failure to downregulate cyclin D1 Dasatinib chemical structure overexpression in melanocytic cells probably promotes cell proliferation and prevents differentiation [29]. Cyclin D1 is a nuclear protein encoded by the CCND1 gene, which is located at chromosome 11q13. CCND1 amplification has been detected in over 44% of acral lentiginous melanomas, but much less frequently in other melanoma subtypes [27] and [31]. All melanoma cases with an increased number of CCND1 copies overexpress cyclin D1. However, about 25% of melanomas that overexpress cyclin D1 have been found to have a normal number of CCND1 copies, suggesting that cyclin

D1 levels are modulated by multiple mechanisms [4], [18] and [27]. It is possible that cyclin D1 overexpression is induced by a defect in its degradation that increases its stability. Cyclin degradation is normally check details regulated by ubiquitin-dependent proteolysis [19] and [24]. Different ubiquitin-dependent proteolytic pathways use enzymes conjugated to different structurally similar ubiquitins. These, in turn, Mirabegron are associated with recognition subunits of proteins targeted by a particular degradation sign. The enzyme that, when conjugated, adds ubiquitin to a lysine residue of a target protein and then, subsequently, adds a series

of additional ubiquitins, forms a polyubiquitin chain that is recognized by a specific receptor protein in proteasomes [1] and [14]. Polyubiquitin chains are linked covalently to the target protein through a cascade of three enzymes: ubiquitin-activating enzymes (E1), ubiquitin-conjugating enzymes (E2), and ubiquitin-protein ligases (E3). In the last stage of this cascade, the ubiquitin-protein ligase (E3) acts as a central component of the ubiquitination pathway, catalyzing the final transfer of ubiquitin from E2 to the substrate [8], [9], [11], [14] and [15]. The interaction of the E2 and E3 proteins is through protein fragments called RING finger proteins. The SCF protein (SKP1-CUL1(CDC53)-F-box) and the Anaphase Promoter Complex (APC) are the two major ubiquitin-ligase complexes.

Z użyciem tabeli randomizacyjnej pacjentów losowo przydzielano do

Z użyciem tabeli randomizacyjnej pacjentów losowo przydzielano do jednej z trzech grup, w których stosowano odpowiednio: 1. 2% żel Lignocainum Hydrochloricum U (producent Przedsiębiorstwo Farmaceutyczne JELFA S.A. Jelenia Góra), czas aplikacji 15 min; Niezaangażowana w pobieranie krwi pielęgniarka wybierała do nakłucia żyłę łokciową, prawej lub lewej ręki, a następnie

aplikowała odpowiednią dawkę preparatu. W każdym find more przypadku zastosowaną warstwę środka znieczulającego pokrywano przezroczystym opatrunkiem okluzyjnym. Zarówno dziecko, jak i osoba pobierająca krew nie wiedziała, do której grupy został przydzielony pacjent. Po upływie wymaganego czasu aplikacji usuwano opatrunek i w warunkach gabinetu zabiegowego pobierano przez nakłucie żyły krew do zleconych badań diagnostycznych. Każdorazowo zabieg pobierania krwi wykonywała ta sama pielęgniarka. W przypadku pojawienia się problemów z jednorazowym pobraniem krwi (np. pęknięcie naczynia), rezygnowano z dalszego uczestniczenia dziecka w badaniu. Po pobraniu krwi i opuszczeniu gabinetu zabiegowego, w wyznaczonym miejscu – sala pobytu dziennego – dziecko otrzymywało kartę z Obrazową Skalą Oceny Bólu (FSP) i zaznaczało rysunek, który odpowiadał nasileniu odczuwanego bólu w trakcie całego zabiegu. Średnie różnice natężenia bólu w trzech badanych grupach porównywano z użyciem

jednoczynnikowej analizy wariancji dla grup przekrojowych ANOVA (test F), a następnie dla zmiennych ciągłych obliczono średnią różnicę między badanymi grupami. Dla zmiennych Wilson disease protein dychotomicznych Palbociclib nmr obliczono ryzyko względne, które definiowano jako

iloraz prawdopodobieństwa wystąpienia danego skutku w grupie eksperymentalnej, w której zastosowano interwencję i tego prawdopodobieństwa w grupie kontrolnej. Wyniki przedstawiono w postaci średniej wraz z 95% przedziałem ufności. Do statystycznej analizy danych użyto komputerowego programu Statistica wersji 5,0, firmy Stat Soft. Analiza wyników została dokonana w grupach wyodrębnionych zgodnie z zaplanowanym leczeniem (ITT – intention to treat analysis). Badanie prowadzone było na Oddziale Pediatrycznego Szpitala Zachodniego im. Jana Pawła II w Grodzisku Mazowieckim w okresie od kwietnia 2004 r. do marca 2005 r. Wstępnie zakwalifikowano 83 pacjentów przyjętych na oddział celem rozszerzenia diagnostyki z zakresu chorób układu oddechowego, alergii, niedoborów masy ciała i wzrostu, zaburzeń ze strony układu pokarmowego, moczowego oraz po wcześniejszym omdleniu i/lub utracie przytomności. Pięć osób (dzieci i/lub opiekunów) po informacji, że czas aplikacji preparatu może wynosić do 1 godziny nie wyraziło zgody na dalsze uczestnictwo w badaniu. Pozostałych 78 pacjentów zgodnie z listą randomizacyjną zakwalifikowano do jednej z 3 interwencji (2% lignokaina, krem EMLA lub placebo), po 26 dzieci w każdej grupie.

From −110 to +10 mV we observed the

normal decrease of th

From −110 to +10 mV we observed the

normal decrease of the Af and As components, but also the increase of the Ass component as shown in the inset to Fig. 2 upper-right (see under Nav1.1 panel). The steady-state inactivation resulted to have a sigmoidal voltage-dependent curve, which, differently from control curves, was characterized by a Daporinad datasheet non complete inactivation and a pedestal at depolarized potentials. This last Ass effect indirectly and strongly affected the window current (normally negligible in control), thus producing a small and not always significant left-shift of the activation curves. When necessary the dose-response relationships of As and Ass components were computed in Fig. 4. In order to visualize the 3D structure of each studied toxin, models of CGTX-II, δ-AITX-Bcg1a and δ-AITX-Bcg1b were constructed using the SWISS-MODEL structure homology-modeling server (http://swissmodel.expasy.org/workspace/) [3]. The tri-dimensional structure of Anthopleurin-A toxin (determined by NMR) [24] (PDB ID: 1ahl) was employed as a template for all models.

The structures were Selleckchem 3 Methyladenine drawn and visualized by DeepView/PDB viewer [12] (http://www.expasy.org/spdbv, version 4.0.1) and PyMOL (The PyMOL Molecular Graphics System, Version 1.2, Schrödinger, LLC., http://www.pymol.org), and were rendered by PovRay (version 3.6 by Persistence of Vision Raytracer, Pty., Ltd.). All the three models were validated by the tools Anolea, DFire, QMEAN, Gromos, Promotif and ProCheck, available in the “structure assessment” tool of the SWISS-MODEL structure homology-modeling server. The toxins employed in this study were obtained according to the previously described procedures [35] and [36]. Considering an urgent need for standardization of nomenclature of animal toxins, especially in sea anemones, we employed a rationale recently suggested [17]

for the novel components eluted during RP-HPLC at 30.24 and 30.57 min from the neurotoxic fraction of B. cangicum venom. ioxilan These peptides were named as δ-Actitoxin-Bcg1a (δ-AITX-Bcg1a) and δ-Actitoxin-Bcg1b (δ-AITX-Bcg1b), respectively. This rationale follows the biological effects exerted by the toxin (δ letter for toxins that delay the inactivation process of ion channels) and the family of the organism which the toxin is derived (“Actitoxin” for toxins isolated from sea anemones of the “Actiniidae” family). Also, full sequences of each peptide were determined in this work. Both sequences were deposited at Uniprot server (http://www.ebi.ac.uk/uniprot/) and their accession numbers were assigned as P86459 and P86460, respectively. As CGTX-II (Uniprot ID: P0C7P9) had already been published [35], we have not changed its name.

Therefore, it has been speculated that the number of pins and are

Therefore, it has been speculated that the number of pins and area of stimuli, similar to the increased amplitude of an S1 response with the increase of intensity of ES, influence the SEFs elicited by MS. It is thus worthwhile to examine the relationship between the conditions of life-like tactile stimuli and cortical activities. In clinical practice, two-point discrimination has been used extensively to evaluate the severity of peripheral nerve injuries

(Jerosch-Herold, 2005 and Lundborg and Rosen, 2004). However, the relationship between the inter-pin distance of 2-pins and S1 activity remains unclear. It is thus important to investigate 5 FU the effect of the number of stimulus pins or inter-pin distance on S1 activities, before two-point discrimination is increasingly used clinically or in research. The present study was designed to investigate the effect of the number of stimulus pins or inter-pin distance of 2-pins on SEF response following MS in the S1 area contralateral to the stimulation. We measured SEFs following the use of a varying number of pins and the inter-pin distance for MS applied to the index finger of healthy participants. Following several different intensities of ES to the index finger, SEF was recorded in order to compare

S1 activity following MS. The typical whole-scalp SEF waveforms detected after MS using 4-pins and 8-pins in a representative subject are shown in Fig. 1. We confirmed a number of deflections in SEF waveforms following MS around the primary sensorimotor area contralateral to the stimulated side. The most APO866 molecular weight prominent SEF deflection was identified approximately

50 ms after MS and the equivalent current dipoles (ECDs) were estimated at the S1 in all subjects. Fig. 2 shows that the representative ECD location estimated at the most prominent deflection after MS with 8-pins superimposed onto a subject′s magnetic resonance image (MRI). The mean ECD locations on axial, coronal, and sagittal planes are summarized in Table 1. There were no significant differences in ECD locations among the five types of stimulus pin numbers (p>0.1). The time courses of the averaged source activities across subjects elicited by each MS with 1-, Loperamide 2-, 3-, 4-, and 8-pins are superimposed and presented in Fig. 3a. We observed a number of deflections in the source activities in all subjects. Each deflection peaked at approximately 28 ms (N20m), 54 ms (P50m), and 125 ms (N100m), and each component could be observed in 6, 12, and 12 out of the 12 subjects, respectively. Table 2 shows the peak latencies of source activities following MS with 1-, 2-, 3-, 4-, and 8-pins. There were no significant differences in peak latencies among the five types of stimulus pin numbers for each component (p>0.05). The source activities for P50m and N100m were significantly altered by a change in the number of stimulus pins (p<0.01, Table 3).

26 and 27 Therefore, the

26 and 27 Therefore, the selleck kinase inhibitor first aim of this cross-sectional study is to verify if there is a tendency

towards an increase in pathogen frequency from peri-implant health to established peri-implant diseases, as previously observed from healthy to diseased periodontal conditions. The second aim of the present study is to test if bacterial frequency is comparative between equivalent periodontal and peri-implant clinical statuses, i.e. healthy peri-implant vs. healthy periodontal sites, mucositis vs. gingivitis and, peri-implantitis vs. periodontitis. This research protocol was reviewed and approved by the Institutional Ethics Committees from University of Taubaté (2008/0098) and Guarulhos University (09/2005). After verbal and written AZD2014 research buy explanations, individuals who agreed to participate signed an informed consent form. Participants received oral hygiene instructions and dental treatment according to their individual needs. This convenience sample population was composed of subjects selected, from January 2006 to June 2010, according

to six specific diagnoses: peri-implant (n = 53 subjects) or periodontal health (n = 53 subjects); peri-implant mucositis (n = 50 subjects) or gingivitis (n = 50 subjects); peri-implantitis (n = 50 subjects) or chronic periodontitis (n = 50 subjects). Eligible subjects were screened from two Clinical Centres, Department of Dentistry of the University of Taubaté and however Department of Periodontics of the University of Guarulhos, according to the following inclusion criteria: male or female; aged between 26 and 52 years; at least fifteen natural teeth; at least one single titanium implant (MKIII, Nobel Biocare) under function for at least one year (for the implant groups). In addition, some exclusion criteria were considered: smoking (current smokers and former smokers); alcohol abuse; diabetes mellitus; immunosuppressive systemic conditions; pregnancy

and lactation; extensive fix or removable orthodontic or prosthetic appliances; local or systemic antibiotic therapy within 6 months prior to biofilm sampling; daily regular use of mouthwash two months prior to the study; any type of periodontal treatment in the past 12 months (for periodontal groups). Clinical parameters were measured by two trained and calibrated examiners at six sites per tooth or implant using a manual periodontal probe (Hu-Friedy PCPUNC 15 Mfg Co. Inc., Chicago IL). After 7 days, periodontal examinations of 10 subjects were repeated showing intra and inter-examiners reproducibility scores higher than 0.85 (Kappa Test) for probing depth (PD) and clinical attachment level (CAL). Intra-class correlation tests showed scores higher than 0.90.

The water holding capacity of the atmosphere is governed by the C

The water holding capacity of the atmosphere is governed by the Clausius-Clapeyron equation, which states that the saturation vapour pressure grows with temperature (at the rate of 6–7% per 1◦C increase in temperature). In other words, warmer air can contain more water vapour. A statistically PD-166866 in vivo significant increase in the frequency of intense precipitation has already been observed at many (but not all) meteorological stations, both

in Europe (Zolina 2012) and in Poland. Moreover, the structure of the precipitation process has changed: short, isolated precipitation events are now giving way to longer precipitation events (Zolina 2012). The mean annual and seasonal precipitation

TSA HDAC has been observed to increase at most weather stations in Poland and to decrease at some others, but many of these changes are not statistically significant. There has been a pronounced, but not ubiquitous, increasing tendency in the intensity of rainfall. However, the inter-annual variability of precipitation is very strong. Changes in the seasonality of precipitation involve a decrease in the ratio of warm-season precipitation to cold-season precipitation (Pińskwar 2009) and also in the proportion of liquid to solid precipitation in winter. The frequency Benzatropine of synoptic weather patterns that are likely to lead to intense precipitation and floods has been on the rise (Niedźwiedź

et al. 2014). There has been an increasing number of local floods in urban areas (flash floods), including large towns (or parts thereof), caused by intense rainfall, when the capacity of the urban sewage systems is too small, or when the urban outflow is obstructed by a flood wave in the river. Flood damage potential in Poland has increased considerably, in the wake of urbanisation and the ubiquitous increase of wealth. Increasing flood exposure results from human encroachment onto floodplains and the economic development of flood-prone areas. The assets at risk from flooding are high, and growing dynamically. Sensitivity to floods has increased since the change of the political and economic system in the early 1990s, accompanying the constantly (for over 20 years now, including the difficult year 2009) growing national GDP. Trends established for Polish tide gauge stations show that the annual mean sea level has been increasing over the last century. Observations of sea level changes in Świnoujście belong to the longest series of records, globally (Pruszak & Zawadzka 2008). More recently the sea level rise has accelerated, up to 0.3 cm yr− 1.

75ρ>0 75), which emphasizes the (expected) assymmetric contributi

75ρ>0.75), which emphasizes the (expected) assymmetric contribution to waves at a certain location by different phases of a certain atmospheric pattern. This is also associated with larger PSS values, especially for coastal locations like Barcelona Erastin datasheet or Valencia (see Fig. 8), lower model biases ( Fig. 9), and smaller absolute RE values along the Catalan coast. However, the improvement in model performance from Setting 2 to Setting 3 is much smaller than that from Setting 1 to Setting

2, which is reflected in all skill measures. The next significant improvement is achieved by the inclusion of the lag-1 dependent variable Hs(t-1)Hs(t-1), i.e., the term ΔtΔt in Eq. (2), as a predictor to predict Hs(t)Hs(t) (Setting 4). The average ρ   score is now around 0.85, with values around 0.9 being seen at many locations ( Fig. 7). This is also associated with greater model skill (larger PSS values) and lower biases (FBI values to closer to unity; see

solid red curves in Fig. 8 and Fig. 9). The average RE (in absolute value) along the Catalan coast is 4.3% for the median HsHs, 14% for the 95th percentile, and 16% for the 99th Dabrafenib percentile, which is reasonably good in the context of HsHs prediction. Being the most complex model among the first group of model settings, Setting 5 includes the term ΔswΔsw as defined in Section 4.2 to further improve representation of swell waves. As summarized earlier, Setting 5 performs the best among Settings 1–5, although the improvement over Setting 4 is small in general. In fact, the small difference between the results of Settings 4 and 5 might be explained by the relatively Uroporphyrinogen III synthase short fetches of the study area and, consequently, the small impact of assuming no time lag δδ between the

origin of swell waves and their propagation to the point of interest as in Settings 3 and 4. In the open ocean where fetches are considerably larger, the difference might be more remarkable. Along the Catalan coast, the improvement of Setting 5 over Setting 4 is more noticeable. As shown in Fig. 8 and Fig. 9, Setting 5 is more skillful than Setting 4 in predicting smaller waves, although it is comparable to Setting 4 for predicting higher waves. Compared to Setting 4, the average absolute RE decreases by 4%, 55% and 50% for, respectively, the 50th, 90th, and 99th (see the dashed red curves in Fig. 10). Thus, we choose to focus on Setting 5 in the subsequent analysis. The second group of model settings (Settings 6–8) are compared in Fig. 11, Fig. 12, Fig. 13 and Fig. 14. They involve the same set of potential predictors as does Setting 5, but with a transformation applied to the wave heights HsHs and/or the squared SLP gradients G0G0, to explore the effect of transforming the non-negative data on model performance, as explained in Section 4.4.