2) There was evidence for an association between the C allele of

2). There was evidence for an association between the C allele of rs9594759 and slower chair rise times (p = 0.04). There was evidence for an association between the C allele of rs9594759 and poorer standing balance (p = 0.04), although this effect was only seen in females with some evidence for a sex difference (p = 0.05 for heterogeneity between males and females, Fig.

S2). There was evidence for heterogeneity between males and females for the association between rs3815148 (COG5) and standing balance, (p = 0.012, Fig. Fulvestrant chemical structure S3) with the observed effects in opposite directions. No other genotypic associations with physical capability measures or evidence for sex differences were observed. Additional adjustment for alcohol consumption for the genotypic effects of rs9594759 did not substantially affect its associations with chair rises (pooled beta for z-score = − 0.031, 95% CI: − 0.060 to − 0.002, p = 0.04, n = 8184) and standing balance in females IDH inhibitor clinical trial (pooled OR = 0.85, 95% CI: 0.75–0.96, p = 0.01, data not shown). In only a relatively small number of tests did the full genotype model represent a significantly better fit than the per allele model: rs9594759 for weight and BMI in Boyd Orr, smoking

status and timed walk in LBC1921; rs2941740 for smoking status in ELSA, socio-economic position in NSHD and balance in CaPS. In this large, multi-cohort study of older adults we investigated associations between robust genetic markers of serum calcium, bone mineral density and osteoarthritis risk and measures of physical capability in six UK cohorts of 12,836 adults aged between 52 and 90 + years. We found marginal evidence for an association between rs1801725 (CASR) and grip strength, with carriers of the allele

associated with raised serum calcium Amobarbital levels, identified from GWAS [19] and [20], having lower grip strength. However, the effect size was small at − 0.03 z-score units for carriers of the T allele, adjusting for age and sex, representing 0.33 kg assuming a standard deviation of 11. We also found some evidence for the association of the BMD-raising allele (C) of rs9594759 (RANKL) [32], [33] and [34] with slower chair rise times and poorer standing balance. This direction was unexpected; however, the interpretation of these results should be treated with caution as the HWE condition was not met for rs9594759 (RANKL) in NSHD and CaPS, and whilst exclusion of the studies is not recommended [59], both studies contributed to the meta-analysis for standing balance and NSHD also contributed to that for chair rises. There were no observed associations with the physical capability measures for the BMD-raising allele of rs2941740 (ESR1).

3) On the

other hand, cyclin D1 expression was <25% in G

3). On the

other hand, cyclin D1 expression was <25% in Groups 1, 2, and 3, but >50% in Group 4 (70.6% of the samples). Group 2 showed no cases with >75% of the cells expressing cyclin D1. A significant negative correlation was observed between ROC1 and cyclin D1 expression levels regardless of neoplasia type (benign or malignant) (p = 0.0008985). Comparisons between ROC1 and cyclin D1 expression in melanomas and melanocytic nevi are shown in Table 1 and Table 2, respectively. In some cases of melanoma, areas with >75% of the cells expressing ROC1 and <25% of cells expressing cyclin D1 were observed adjacent to areas wherein ROC1 was positive in <25% of the cells, and cyclin D1 was expressed in >75% of the cells. This was found to be independent of increased gene expression (Fig. 4). The ROC1/cyclin D1 relationship did not vary with age, gender, or lesion site in either melanomas or melanocytic nevi (p > 0.05).

Increased C646 cost ROC1 protein expression, as compared with cyclin D1 expression, RGFP966 ic50 predominated in all samples (65% of cases; n = 78). In the melanocytic nevus group, the ROC1 expression increase was remarkably predominant in relation to cyclin D1 expression (86.2% of the cases). In melanomas, this ROC1 expression predominance was also observed, but in only 45.2% of the cases (p < 0.001) ( Table 3). Although ROC1 and cyclin D1 expression levels were predominantly proportional in melanomas with thickness >2 mm, and although a great number of cases with melanomas >4 mm (35.3%) showed increased cyclin D1 expression in comparison with ROC1 levels, no statistically significant difference was seen among the groups (p = 0.166). Only in the acral lentiginous melanoma group was cyclin D1 expression greater than that of ROC1 in a large number of cases (40%). On the other hand, this group also showed the largest number of cases with increased ROC1 expression as compared TCL to cyclin

D1 expression (50%). No statistically significant difference in the ROC1/cyclin D1 relationship was observed in relation to melanoma histological type (p = 0.605). Six cases (five melanomas and one melanocytic nevus) exhibited CCND1 gene amplification. In two amplified cases, one was acral lentiginous melanoma and the other was nodular melanoma with Breslow thickness of >4 mm. Cyclin D1 was expressed in 51–75% of the acral lentiginous melanoma cells and in >75% of the nodular melanoma cells. In both the acral lentiginous and nodular melanomas, ROC1 expression was present in <25% of the cells. In the other amplified melanomas (2 SSM and 1 LMM), in one case, the Breslow's thickness was <1 mm, in another it was 1.01–2 mm, and in the other it was 2.01–4 mm. Of these three amplified melanomas, two showed cyclin D1 and ROC1 expression in 51–75% of the cells, while in the other case, cyclin D1 positivity was <25%, and ROC1 was expressed in >75% of the cells.

Once sample has been acquired, then the end of the needle is seal

Once sample has been acquired, then the end of the needle is sealed and the needle body is inserted into the hot injector

selleck kinase inhibitor of the gas chromatograph. Water collected with the sample is in this case an advantage as the pressure change associated with its vaporization is used to drive the VOC into the column. Sensitivity can be increased simply by increasing the sample volume, until breakthrough occurs (Trefz et al., 2012). Needle trap methods provide a simple, robust, high sensitivity and low cost alternative to presently used seawater sampling methods (Alonso et al., 2011a, Bagheri et al., 2011 and Risticevic et al., 2009). Here, we exploit the suitability of needle trap devices for the study of VOCs in seawater samples. A sampling method based on purging volatile tracers out of water samples directly onto the needle traps has been developed and evaluated for DMS, isoprene, benzene, toluene, p-xylene,

(+)-α-pinene and (−)-α-pinene. Subsequently the method was applied in a CO2 enrichment field study. Seawater concentrations of dimethyl sulfide (DMS), isoprene and monoterpenes were monitored from May 8 to June 6, 2011. Datasets of DMS and isoprene during this period are presented here. These examples show contrasting responses upon ocean acidification. In the field, additional method validation was achieved for DMS through an inter-laboratory comparison Angiogenesis inhibitor between our NTD GC–MS method and an independent purge and trap technique using gas chromatograph–flame photometric analysis (P&T GC–FPD). Commercial side-hole NTDs (needle trap devices) consisting of a 23-gauge, 60 mm long stainless steel needle, packed with 1 mg polydimethyl siloxane (PDMS), 0.4 mg Carbopack X and 0.5 mg Carboxen

1000 (1 cm each), were purchased from PAS Technology, Magdala, Germany (Fig. 1). Gas entering the needle trap was directed over the weaker adsorber first (PDMS). Prior to first use, the NTDs were conditioned in the gas chromatograph injection port at 300 °C for 30 min under a permanent helium flow (1 ml/min) to remove impurities. Gas tight syringes, glass fiber filters (25 mm, Whatman GF/F) and water sampling syringes (10 ml) were purchased from Sigma Aldrich. A commercial Baricitinib multi-component gas standard mix (Apel-Riemer Environmental Inc.) was used for calibrations (stated accuracy 5 %). Helium 6.0 and synthetic air (20.5 % O2, rest N2, hydrocarbon free) were from Westfalen AG, Germany. A sampling set up (supplied by PAS Technology) comprising of a mass flow controller (5–250 ml/min, calibrated on He), vacuum pump, voltage regulator, temperature regulator, purge tube heating body and a manual water inlet kit was used to extract VOCs from water samples. The set-up is shown schematically in Fig. 2. Glass purging tubes (10 ml sampling volume) including a bottom frit were prepared in the glass workshop of the Max Plank Institute in Mainz.

In estimating the net atmospheric flux to sea areas one should no

In estimating the net atmospheric flux to sea areas one should note that in the 1990s many fluxes (CO2, NH3) over the sea surface were found to be bidirectional and that deposition should be estimated by a coupled marine-atmospheric model. The effects of European international shipping on the basis of countryby-country deposition and ozone concentrations have been studied in Jonson et al. (2000). Deposition to the BS caused by European countries and sea traffic is reported annually in EMEP source-receptor matrices. A review of existing studies on the impacts of shipping emissions of different chemical compounds on air quality in coastal areas

is presented and discussed in detail in EEA (2013), along with a summary of the results over the area considered, methodological data and conclusions. selleck kinase inhibitor The nitrogen deposition to the BS was selleck chemicals calculated with the Hilatar chemistry-transport model (Hongisto 2003). As input, the model uses the forecasts of the FMI operative HIRLAM hydrostatic weather prediction model (HIgh Resolution Limited Area Model, Unden et al. 2002). The Hilatar, a dynamic Eulerian model covering Europe with a zooming model over the Baltic

Sea and its close surroundings (the BS model with 0.068 deg resolution), provides gridded estimates of the fluxes and concentrations of oxidised and reduced nitrogen and sulphur compounds. Gaseous (g) and particle (p) concentrations are calculated for the following substances: NOx(g), HNO3(g), NO3(p), PAN(g), NH4NO3(p), NH3(g), SO2(g), SO4(p) and (NH4)1.5SO4(p), where PAN is peroxyacetyl nitrate and NOx = NO + NO2. The chemistry module comprises the EMEP-MSC-W chemistry code ( Iversen et al. 1989) with some modifications ( Hongisto 2003). The model does not have ozone as a variable, because in photooxidant codes the main radical concentrations influencing the chemical transformation of nitrogen and sulphur chemistry are calculated inside the model. Their values are, Anacetrapib however, rather seldom verified or even presented. For basic

acid chemistry one can use measurement-based functions for all radicals and oxidants needed. The Hilatar model, run since 1993, has the HIRLAM grid of the current operative model: horizontally rotated spherical coordinates and vertically hybrid sigma coordinates with selected (now 21) layers up to 5–10 km in height. The long-range transported compounds at the borders of the BS model domain, calculated by the 0.15° resolution European-scale model, are included in the advected air with six hour intervals. For the years 2008–2011, both models used the HIRLAM version V71 vertical grid; from the 60 available vertical levels the 18 lowest (up to around 1.5 km) and three additional levels (at around 2 km, 2.8 km and 5.1–5.3 km) are used. In Hilatar, horizontal advection is solved numerically according to Bott’s (1989) method, while chemistry uses the Hesstvedt et al.

Meanwhile, genomic data from 27 diverse maize inbred lines showed

Meanwhile, genomic data from 27 diverse maize inbred lines showed that the genome consists of highly divergent haplotypes with 10- to 30-fold variations in recombination rates [27]. This reinforces the concept that maize is a highly polymorphic species. However, it also shows that there are often large genomic regions that have little or no variation [28]. Much valuable LDE225 in vivo information likely underlies the genome

signature due to selection that can be exploited for breeding. The objectives of this study were to (i) confirm the genetic locus for cob glume color using a genome wide association study (GWAS) with high resolution SNPs, (ii) reveal the genome pattern surrounding it, and (iii) find find more evidence of the effects of selection across the target region. The results reported here may provide insights as to the manner by which breeding efforts have affected and will affect genome evolution.

A set of 283 diverse inbred lines, representing the modern temperate maize elite inbred lines in China [29] and [30], was used for genotyping with 55,000 SNPs and GWAS. Forty of the lines from this association panel and 47 tropical lines with white cob glumes were re-sequenced 10 × through an international collaboration (Xu et al., in preparation). These plant materials were grown in Sanya, Hainan province (18°45°N, 109°30°E), during the winter of 2011–2012. Each line was planted in a plot with 20 plants in a 4.5 m row with 0.6 m spacing between rows. Normal agronomic practices were used in field management. After harvest, cob glume color was

scored for each line as “0” for white and “1” for other colors. The scores were used for GWAS. Based on the B73 reference sequence, 56,110 evenly spaced SNPs were featured on the MaizeSNP50 BeadChip (Illumina, Inc.). These were selected from several public and private sources and included 984 negative controls. DNA was extracted from the 283 temperate lines by a modified CTAB procedure Bcl-w [31]. Before genotyping, each DNA sample was evaluated using gel-electrophoresis and spectrophotometry (NanoDrop 2000, Thermo Scientific). As controls, four lines (Qi 319, Huangzao 4, Ye 478 and Dan 340) were added to each of the six independent BeadChips. SNP genotyping was performed using the MaizeSNP50 BeadChip by Emei Tongde (Beijing, China). SNP calling for the 283 samples was implemented according to the Infinium HD Assay Ultra Protocol Guide (Illumina, Inc.). After filtering out monomorphic and non-specific SNPs, a subset of 44,235 SNPs with known physical positions was generated, with an average heterozygosity of 0.5%. Within the four controls, the mean reproducibility between replicates across all data points was 99.9%, which is consistent with high-quality data for replicates of the B73 maize line using the same chip (Illumina, Inc.). The error rate (ER) for genotyping was 0.0353%.

Each factor level was selected based on preliminary studies Prel

Each factor level was selected based on preliminary studies. Preliminary results from a full factorial design had shown significant curvature (data not shown), hence a central composite design was chosen, in particular, a ‘face

centered’ design as only two types of extruded biomass were available (7% and 80% xylose removal). The ratio of the total amount of glucose produced in the hydrolyzate to the total theoretical amount of glucose in the steam-exploded corncobs (analyzed after acid hydrolysis) was chosen as the response for analysis. The experimental design was developed using the software Design Expert, version 8.0.7.1 (Stat Ease, selleck inhibitor Inc. USA). The resulting 22 experimental conditions, as well as three center point replicates for each type of biomass, were tested in triplicate and data is presented as the average of triplicates ± standard deviation. All experiments were performed fully

randomized, and the data was fitted via linear regression to a second order model: equation(2) y=β0+Σi=1kβixi+Σi=1kβiixi2+Σ1≤i≤jkβijxixj+ϵWhere y is the predicted response, xi represents the independent variables, k is the number of variables, β0 is the interception coefficient, βi represents the linear coefficient of each independent variable, βii represents the coefficients Seliciclib of the quadratic terms, βij represents the coefficients of the interaction effects and ε is the random error. Analysis of the variance (ANOVA) was performed and the significance of each variable, the interaction, and quadratic effects were determined

based on a significance of α = 0.05 using the F -test. The fitted model was evaluated by R2, adjusted R2, adequate precisior and the lack of fit coefficient for determining the adequacy. In addition, the fitted model was validated by performing experiments using the identified conditions of the significant variables [1]. The carbohydrate composition of the investigated corncobs before and after steam explosion and after different extruder treatments was measured after acid hydrolysis [9], [21] and [5]. N-acetylglucosamine-1-phosphate transferase The data are shown in Table 1 (based on total dry matter). The relative glucose content, which was the largest fraction of monosaccharides, increased from 41% to 66% and 58%, respectively, depending on different extrusion process conditions. The hemicelluloses fraction was largely hydrolyzed to xylose under high temperature and pressure during the steam explosion pretreatment. 7% xylose removal from the steam exploded corncobs was achieved through the extrusion process at a barrel temperature of 65 °C and a screw speed of 100 rpm without adding water, while 80% xylose removal was achieved when the barrel temperature increased to 100 °C and water was injected at Barrel 8 at 2.9 kg/h. Arabinose, galactose, and mannose were found in minor fractions (<5.0%). SEM images of untreated and extruded corncobs with different xylose removals at different magnifications are shown in Fig. 2.

, 2006) The current EPA RfD of 0 3 μg/kg-day derived from a NOAE

, 2006). The current EPA RfD of 0.3 μg/kg-day derived from a NOAEL for skin lesions in SW Taiwan incorporates an uncertainty factor of 3, based on insufficient data to preclude reproductive toxicity and potential variation in individual sensitivity (EPA, 1993). EPA, however, noted a lack of clear consensus among agency scientists and that strong scientific arguments could support values between 2 and 3 of the RfD (i.e., from 0.1 μg/kg-day to 0.8 μg/kg-day). In evaluating a specific RfD for CVD, however,

other endpoints such as reproductive toxicity (except for effects related to CVD) would not be considered. The SGLT inhibitor available evidence for potential individual differences in sensitivity to arsenic indicates that the Bangladesh population would be more sensitive than the U.S. population PD-0332991 ic50 based on a number of factors. South Asians are reported to be susceptible to coronary artery disease, and Bangladeshis are reported to be even more prone to heart disease, even when living abroad in countries such as the United States or United Kingdom (Islam and Majumder, 2013). In addition to having some of the common risk factors, heart disease may be increased in Bangladeshis by nutritional deficiencies and related conditions (e.g., hyperhomocysteinemia), low birth weight and childhood malnutrition,

high prevalence of betel nut use, and possibly genetic susceptibility (Gamble et al., 2005a, Islam and Majumder, 2013 and Pilsner et al., 2009). Consistent with lower intakes of folate as noted previously, folate biomarkers were lower in Bangladesh than in the U.S. population. Idoxuridine Median plasma/serum folate levels among controls without skin lesions in a subgroup of the HEALS cohort (3.4 ng/mL) (Pilsner et al., 2009) and

in a larger portion of the cohort (4.6 ng/mL in women, 3.7 ng/mL in men) (Gamble et al., 2005a) were considerably lower than in the United States (median in 2005–2006 of 12.2 ng/mL) (McDowell et al., 2008). The prevalence of a low serum folate level (<3 ng/mL) is less than 1% in the U.S. population (McDowell et al., 2008). Although elevated arsenic exposure may also contribute to lower folate levels in HEALS participants, the relatively weak inverse correlation between water arsenic concentration and folate levels (r = −0.13) ( Gamble et al., 2005b), indicates an overall reduced folate intake in Bangladesh relative to the United States. Lack of folic acid fortification of foods in Bangladesh is also compounded by traditional cooking practices involving prolonged cooking, which can oxidize up to 95% of the naturally occurring folate in foods (FAO, 2001 and Gamble et al., 2005a). By contrast, folic acid is much more resistant to oxidation and has nearly 100% bioavailability compared to 25–50% for natural folate in foods (FAO, 2001). In the HEALS cohort, plasma folate levels were correlated with urinary arsenic forms in the expected directions for impairment of arsenic methylation (i.e.

However the experimental biodegradability

However the experimental biodegradability CHIR 99021 cannot be applied to the COD methodology as it was determine from de VS of the. Also the relative error is obtained from the Eq. (8) comparing the BMPexp and BMPthBD. For the COD Eq. (2) the theoretical production (BMPth) follows the same behavior for biological sludge and OFMSW as the experimental results, where higher productivity was achieved by the OFMSW with a COD of 542 g/kg than the biological sludge (77.1 g/kg COD). In the co-digestion mixtures the productivity decreases with the COD content and the co-digestion mixture productivities do not surpass the productivity

of the sole substrates, although when Trametinib mw applying the experimental biodegradability (BMPthBD) the behavior changes, increasing the productivity for all the co-digestion mixtures from the sole substrates as occurs in the experimental results. The highest errors are obtained for this method with agreements lower than 90%. Despite the fact that the theoretical results obtained for the elemental composition equation method follows behavior similar to the previous method and the experimental results, the values are lower, but it gets agreements higher than 90%. However the co-digestion mixtures get similar increases from the sole substrates

OFMSW and biological sludge for co-digestion 1, while co-digestions 2, 3 and 4 increase only from the sole biological sludge. In this case the theoretical productivity decreases in those substrates with higher hydrogen and nitrogen presence, which can produce toxic concentration G protein-coupled receptor kinase of ammonia and hydrogen sulfide [8]. It is also observed that the productivity increases with the rise of the COD and with the increase of the C/N ratio (Table 3). Some researchers have suggested that the C/N ratio for optimum digestion performance is in the range of 20–30, while many have demonstrated

that digestion can be successfully performed using a wider range of C/N ratios [13] and [37]. The organic fraction composition Eq. (5), obtains prediction results with a relative error % below 10%. The productivity increases with the proportion of lipids, as lipids exhibit a much higher biogas potential (1 m3 per kg of volatile solids) than carbohydrates, proteins or cellulose [36], nevertheless their kinetics are slower with higher fiber percentages (Table 4). Applying the biodegradability of the experimental results, none of the co-digestion mixtures exceed the productivity of the sole OFMSW. Otherwise the experimental results showed a different behavior, meaning that the synergistic effects could play an important role in the biodegradability of the co-digestion of these two substrates.

07, p =  289] We found no difference in the average time of cons

07, p = .289]. We found no difference in the average time of conscious intention between GTS patients and controls in our group of adolescents. A previous study had reported a delay in conscious intention in adults with GTS relative to controls (Moretto et al., 2011) but this result was not replicated in our younger and larger sample. The absence of delay in adolescence combined with delayed experience of volition in adults with GTS suggests that adults may learn the experience of volition. In healthy adults, the normal experience of intention prior to voluntary action may CHIR-99021 mouse reflect

prolonged perceptual learning at discriminating the internal signals that characterise volition. Persistent co-occurrence of voluntary and involuntary movement in GTS could make this discrimination problem harder. Therefore, patients with GTS may show delayed learning about their own volition, or may extinguish such learning after it has occurred, as a result of prolonged tic behaviour. Adults have prolonged experience of their own voluntary action, and may have learned the discriminative perceptual markers of volition. However, for an adult with GTS, frequent

Cobimetinib concentration tics may have made this discrimination harder, leading to a more conservative criterion for detecting the signal among noise. GTS adults may thus lack the normal anticipatory awareness of intentional action. In our adolescent sample, the two groups do not yet diverge in this way. That is, we suggest that the delayed experience of volition in adult GTS represents a failure of perceptual learning for volition-related signals, due to masking click here by tics and tic-related factors, such as premonitory urges. Some possible factors are discussed in the next section. GTS is characterised by tics. Our results

showed several influences of ticcing on the experience of voluntary action. These results are consistent with the broad theory that the experience of volition involves learning a perceptual discrimination between the distinctive internal states and signals corresponding to preparation of voluntary actions, and other, involuntary body movements. For example, a striking result of our regression analysis was that subjective experiences linked to involuntary tic movements (measured by the PUTS) provided the single strongest predictor of volition. Participants who experienced strong premonitory urges prior to tics had a later perception of the intention preceding voluntary action. Stronger premonitory urges preceding involuntary movements could impair detection of the distinctive experience of volition, since urges to tic would constitute perceptual noise masking actual intentions.

Although infliximab is indicated for the treatment

of UC

Although infliximab is indicated for the treatment

of UC only as a 5-mg/kg dose regimen, Talazoparib mw for the purpose of these analyses, data from patients who received the 10-mg/kg dose regimen in the ACT-1 and ACT-2 trials were included for a more robust evaluation and interpretation of the concentration-response relationship. Clinical outcomes were assessed using the Mayo score at week 8 (ACT-1 and ACT-2), at week 30 (ACT-1 and ACT-2), and at week 54 (ACT-1 only). Clinical response, defined as a decrease from baseline in the total Mayo score of at least 3 points and at least 30%, and with an accompanying decrease in the rectal bleeding subscore of at least 1 point or an absolute rectal bleeding subscore of 0 or 1, was the primary end point for both the ACT-1 and ACT-2 trials. Clinical remission was defined as a total Mayo score of 2 points or lower, with no individual subscore exceeding 1 point. Mucosal healing was defined by an endoscopy subscore of 0 or 1. For PK evaluations, patients were followed up through week 54 in ACT-1 or through week 42 in ACT-2. In ACT-1, blood samples for determining serum infliximab concentrations were drawn just before and 1 hour after the infusions at weeks 0, 2, 6, 14, and 46, and just before the infusions at

weeks 30 and see more 38. Additional 4��8C blood samples for determination of serum infliximab concentrations were drawn at the week-8 and week-54 nondosing visits (Supplementary Figure 1). In ACT-2, blood samples were drawn

just before and 1 hour after the infusions at weeks 0 and 2, and just before the infusions at weeks 6 and 14. Additional blood samples for serum infliximab concentration analysis were drawn at the week-8, week-30, and week-42 nondosing visits (Supplementary Figure 1). Serum infliximab concentrations were determined using a validated enzyme-linked immunosorbent assay,16 with a lower limit of quantification of 0.1 μg/mL. Of the 484 patients randomized to infliximab (5 or 10 mg/kg) in the ACT-1 and ACT-2 trials, 482 received at least 1 infusion and had appropriate serum infliximab concentration data. ATI were determined using an antigen-bridging enzyme immunoassay.16 Similar to other enzyme immunoassays, this assay was susceptible to drug interference and was not able to detect ATI accurately in the presence of a measurable infliximab concentration. For the purpose of this analysis, patients were classified as positive if ATI were detected in their serum samples at any visit, whereas all other patients were regarded as nonpositive for ATI. The patient population for these analyses included only those who received at least 1 infusion of infliximab at a dose of 5 or 10 mg/kg.