Partitioning Of Label Research Articles

Identification of protein functions in mouse with a label space partition method.

Protein is very important for almost all living creatures because it participates in most complicated and essential biological processes. Determining the functions of given proteins is one of the most essential problems in protein science. Such determination can be conducted through traditional experiments. However, the experimental methods are always time-consuming and of high costs. In recent years, computational methods give useful aids for identification of protein functions. This study presented a new multi-label classifier for identifying functions of mouse proteins. Due to the number of functional types, which were termed as labels in the classification procedure, a label space partition method was employed to divide labels into some partitions. On each partition, a multi-label classifier was constructed. The classifiers based on all partitions were integrated in the proposed classifier. The cross-validation results proved that the proposed classifier was of good performance. Classifiers with label partition were superior to those without label partition or with random label partition.

Multi-label classification and label dependence in in silico toxicity prediction

Most computational predictive models are specifically trained for a single toxicity endpoint and lack the ability to learn dependencies between endpoints, such as those targeting similar biological pathways. In this study, we compare the performance of 3 multi-label classification (MLC) models, namely Classifier Chains (CC), Label Powersets (LP) and Stacking (SBR), against independent classifiers (Binary Relevance) on Tox21 challenge data. Also, we develop a novel label dependence measure that shows full range of values, even at low prior probabilities, for the purpose of data-driven label partitioning.Using Logistic Regression as the base classifier and random label partitioning (k = 3), CC show statistically significant improvements in model performance using Hamming and multi-label accuracy scores (p<0.05), while SBR show significant improvements in multi-label accuracy scores. The weights in the Logistic Regression and Stacking models are positively associated with label dependencies, suggesting that learning label dependence is a key contributor to improving model performance.An original quantitative measure of label dependency is combined with the Louvain community detection method to learn label partitioning using a data-driven process. The resulting MLCs with learned label partitioning were generally found to be non-inferior to their corresponding random or no label partitioning counterparts. Additionally, using the Random Forest classifier in a 10-fold stratified cross validation Stacking model, we find that the top-performing stacking model out-performs the corresponding base model in 11 out of 12 Tox21 labels. Taken together, these results suggest that MLC models could potentially boost the performance of current single-endpoint predictive models and that label partitioning learning may be used in place of random label partitionings.

Planting trees for scalable and efficient canonical hub labeling

Hub labeling is widely used to improve the latency and throughput of Point-to-Point Shortest Distance (PPSD) queries in graph databases. However, constructing hub labeling, even via the state-of-the-art Pruned Landmark Labeling (PLL) algorithm is computationally intensive. PLL further has a sequential root order label dependency that makes it challenging to parallelize. Hence, the existing parallel approaches are often plagued by label size increase, poor scalability and inability to process large weighted graphs. In this paper, we develop novel algorithms that construct the minimal (guaranteed) Canonical Hub Labeling on shared and distributed-memory parallel systems in a scalable and efficient manner. Our key contribution, the PLaNT algorithm, provides an embarrassingly parallel approach for label construction that scales well beyond the limits of current practice. Our approach is the first to employ a collaborative label partitioning scheme across multiple nodes of a cluster, for completely in-memory labeling and parallel querying on massive graphs whose labels cannot fit on a single node. On a single node with 72-threads, our shared-memory algorithm is up to 47.4X faster than sequential PLL. While our labeling time is comparable to the state-of-the-art shared-memory paraPLL, our label size is 17% smaller on average. PLaNT demonstrates superior parallel scalability. It can process significantly larger graphs and construct labeling orders of magnitude faster than the state-of-the-art distributed paraPLL. Compared to the best shared-memory parallel algorithm, it achieves up to 9.5X speedup on a 64 node cluster.

Predicting Drug-Target Interactions With Multi-Label Classification and Label Partitioning.

Identifying drug-target interactions is crucial for drug discovery. Despite modern technologies used in drug screening, experimental identification of drug-target interactions is an extremely demanding task. Predicting drug-target interactions in silico can thereby facilitate drug discovery as well as drug repositioning. Various machine learning models have been developed over the years to predict such interactions. Multi-output learning models in particular have drawn the attention of the scientific community due to their high predictive performance and computational efficiency. These models are based on the assumption that all the labels are correlated with each other. However, this assumption is too optimistic. Here, we address drug-target interaction prediction as a multi-label classification task that is combined with label partitioning. We show that building multi-output learning models over groups (clusters) of labels often leads to superior results. The performed experiments confirm the efficiency of the proposed framework.

Preferential Alternatives to Returning All Crop Residues as Biochar to the Crop Field? A Three-Source 13C and 14C Partitioning Study.

The simultaneous effects of biochar on soil organic matter (SOM, C4) and sweet potato (SP) residue (Ipomoea batatas, C3) mineralization were studied over 180 days via 13C and 14C isotopic label partitioning. Upon concomitant SP residue addition, biochar mineralization decreased by 11% of the total added biochar-C. Compared to positive priming effects induced by biochar amendment alone on SOM (0.46 mg C g-1 soil) at 180 days, amendment solely with SP residues induced significantly larger effects (1.5 mg C g-1 soil). Combination biochar and SP residue addition reduced SOM mineralization by 20.5% and increased SP residue mineralization by 10.1%. Biochar addition caused preferential uptake of SP residues over SOM by microbes. Thus, the lower priming effects on SOM and CO2 emission induced by biochar amendment with or without SP residues compared to that from SP residue addition alone may result in crop residues being partly pyrolyzed to biochar in the cropland.

Enhanced Trifluralin Metabolism Can Confer Resistance in Lolium rigidum.

Resistance to the pre-emergence herbicide trifluralin is increasing in Australian annual ryegrass ( Lolium rigidum) populations. Three L. rigidum populations (R1, R2, and R3) collected from Australian grain fields were identified with trifluralin resistance. Both target-site and nontarget-site resistance mechanisms were investigated. No target-site α-tubulin mutations were detected in populations R1 and R3, while an Arg-243-Lys mutation was found in R2. Compared with the three trifluralin-susceptible populations, enhanced [14C]-trifluralin metabolism, quantified by measuring the amount of [14C] label partitioning into the polar phase of a hexane:methanol system, was identified in all the three resistant populations. This is the first report of metabolic resistance to trifluralin. Coevolution of target-site and nontarget-site resistance to trifluralin is occurring, and metabolic resistance is not rare in L. rigidum populations in Australia. A method was established for trifluralin metabolic resistance detection, overcoming the difficulties of quantifying this highly volatile herbicide by chromatographic methods.

Plant and Microbial Uptake and Allocation of Organic and Inorganic Nitrogen Related to Plant Growth Forms and Soil Conditions at Two Subarctic Tundra Sites in Sweden

ABSTRACTIn order to follow the uptake and allocation of N in different plant functional types and microbes in two tundra ecosystems differing in nutrient availability, we performed a 15N-labeling experiment with three N forms and followed the partitioning of 15N label among plants, microorganisms and soil organic matter. At both sites the deciduous dwarf shrub Betula nana and the evergreen Empetrum hermaphroditum absorbed added 15N at rates in the order: NH4 + > NO3 − > glycine, in contrast to the graminoid Carex species which took up added 15N at rates in the order NO3 − > NH4 + > glycine. Carex transported a high proportion of 15N to aboveground parts, whereas the dwarf shrubs allocated most 15N to underground storage. Enhanced 13C in Betula nana roots represents the first field evidence of uptake of intact glycine by this important circumpolar plant. Plant and microbial uptake of label was complementary as plants took up more inorganic than organic N, while microbes preferred organic N. Microbes initially took up a large part of the added label, but over the following four weeks microbial 15N decreased by 50% and most 15N was recovered in soil organic matter, while a smaller but slowly increasing proportion was retained in plant biomass.

Fluxomics: mass spectrometry versus quantitative imaging

The recent development of analytic high-throughput technologies enables us to take a bird's view of how metabolism is regulated in real time. We have known for a long time that metabolism is highly regulated at all levels, including transcriptional, posttranslational and allosteric controls. Flux through a metabolic or signaling pathway is determined by the activity of its individual components. Fluxomics aims to define the genes involved in regulation by following the flux. Two technologies are used to monitor fluxes. Pulse labeling of the organism or cell with a tracer, such as 13C, followed by mass spectrometric analysis of the partitioning of label into different compounds provides an efficient tool to study flux and to compare the effect of mutations on flux. The second approach is based on the use of flux sensors, proteins that respond with a conformational change to ligand binding. Fluorescence resonance energy transfer (FRET) detects the conformational change and serves as a proxy for ligand concentration. In contrast to the mass spectrometry assays, FRET nanosensors monitor only a single compound. Both methods provide high time resolution. The major advantages of FRET nanosensors are that they yield data with cellular and subcellular resolution and the method is minimally invasive.

Contribution of explant carbohydrate reserves and sucrose in the medium to bulb growth of lily regenerated on scale segments in vitro

Bulb size is an important factor determining phase change in Lilium: phase change only occurs in bulblets over a certain threshold weight. After phase change has occurred, bulblets sprout with a stem with many leaves. Juvenile bulblets sprout with only a few leaves. The factors contributing to bulb size were studied during in vitro regeneration of bulblets on scale segments. The larger the explants, the larger the regenerated bulblets. Explant size influenced bulb growth during the complete culture period. Bulb growth was stimulated by a high sucrose concentration. The contribution of the medium and the explant reserves to bulb growth were studied in large and small explants using labelled sucrose. Sucrose was mainly taken up through the cut surfaces. In freshly cut explants, the rate of uptake was correlated with the size of the contact area, but at later stages, when regenerating organs were present, the difference in uptake rate of small and large explants almost disappeared. Small explants had a larger sink activity than large ones. Explants with regenerating organs took up more sucrose than freshly cut explants. Sucrose uptake and bulb growth were rather constant in the later phases and doubled when the sucrose concentration was doubled. Partitioning of label from the sucrose over the various organs, was also rather constant in time: approximately 25% was accumulated in the bulblets, 35–45% in the explant and 30–35% was converted to CO2. About half of the label in the explant was recovered at the proximal side where regeneration takes place. Sucrose, once incorporated in the storage pools in the explant, either remained in the explant or was converted to CO2. Redistribution to the growing bulblets hardly occurred. The percentage of bulb growth that could be attributed to uptake of medium components was constant over the regeneration period: 45–50% for large and 65–75% for small explants.

Activation of pyrophosphate:fructose‐6‐phosphate 1‐phosphotransferase by fructose 2,6‐bisphosphate stimulates conversion of hexose phosphates to triose phosphates but does not influence accumulation of carbohydrates in phosphate‐deficient tobacco cells

The aim of this work was to investigate the contribution of fructose 2,6-bisphosphate to the regulation of carbohydrate metabolism under phosphate stress. The study exploited heterotrophic tobacco callus lines expressing a modified mammalian 6-phosphofructo-2-kinase/fructose 2,6-bisphosphatase that increased the fructose 2,6-bisphosphate content of the tissue. The phosphate status of two transgenic and one untransformed cell line was perturbed by incubation with 2-deoxyglucose, a phosphate-sequestering agent, and by growth of callus on phosphate-depleted media. 31P-NMR spectroscopy confirmed that both treatments decreased cellular levels of inorganic phosphate and phosphorylated metabolites. Despite large decreases in the amounts of phosphate esters, UDPglucose and adenylates in response to phosphate deficiency, the fructose 2,6-bisphosphate content of each line was unaffected by 2-deoxyglucose and increased during growth on phosphate-limited media. Short-term treatment of callus with 2-deoxyglucose had only minor effects on the carbohydrate status of each line, whereas long-term phosphate deficiency caused an increase in starch and a decrease in soluble sugar content in both transgenic and control lines. There were no consistent differences between the three callus lines in metabolism of [U-14C]glucose in response to incubation with 2-deoxyglucose. In contrast, there was a decrease in partitioning of label into glycolytic products (particularly organic acids) in untransformed callus during growth on phosphate-depleted medium. This decrease was greatly attenuated in the transgenic lines with increased fructose 2,6-bisphosphate content. This suggests that the conversion of hexose phosphates to triose phosphates is constrained under phosphate-deficient conditions, and that this restriction can be relieved by activation of pyrophosphate:fructose-6-phosphate 1-phosphotransferase. However, since the transgenic and control lines did not differ in the extent to which the carbohydrate content changed in response to growth on phosphate-depleted media, it is concluded that an increase in flux through pyrophosphate:fructose-6-phosphate 1-phosphotransferase is not a major component of the metabolic response of heterotrophic tobacco cells to phosphate deficiency.

Regulation of photosynthesis during Arabidopsis leaf development in continuous light.

Previous investigations in our laboratory have shown that leaf developmental programming in tobacco is regulated by source strength. One hypothesis to explain how source strength is perceived is that hexokinase acts as a sensor of carbohydrate flux to regulate the expression of photosynthetic genes, possibly as a result of sucrose cycling through acid invertase and hexokinase. We have turned to Arabidopsis as a model system to study leaf development and have examined various photosynthetic parameters during the ontogeny of a single leaf on the Arabidopsis rosette grown in continuous light. We found that photosynthetic rates, photosynthetic gene expression, pigment contents and total protein amounts attain peak levels early in the expansion phase of development, then decline progressively as development proceeds. In contrast, the flux of (14)CO(2) into hexoses increases modestly until full expansion is attained, then falls in the fully expanded leaf. Partitioning of carbon into hexoses versus sucrose increases until full expansion is attained, then falls. The in vitro activities of hexokinase, vacuolar acid invertase, and cell wall acid invertase do not change until the late stages of senescence, when they increase markedly. At this time there are also dramatic increases in hexose pool sizes and in senescence-associated gene (SAG) expression. Taken together, our results suggest that invertase and hexokinase activities do not control the partitioning of label into hexoses during development. We conclude that our data are not readily compatible with a simple model of leaf development, whereby alterations in photosynthetic rates are mediated directly by hexose flux or by hexose pool sizes. Yet, these factors might contribute to the control of gene expression.

Influence of Etherel and Gibberellic Acid on Carbon Metabolism, Growth, and Essential Oil Accumulation in Spearmint (Mentha Spicata)

Changes in growth parameters and 14CO2 and [U-14C]-sucrose incorporation into the primary metabolic pools and essential oil were investigated in leaves and stems of M. spicata treated with etherel and gibberellic acid (GA). Compared to the control, GA and etherel treatments induced significant phenotypic changes and a decrease in chlorophyll content, CO2 exchange rate, and stomatal conductance. Treatment with etherel led to increased total incorporation of 14CO2 into the leaves wheras total incorporation from 14C sucrose was decreased. When 14CO2 was fed, the incorporation into the ethanol soluble fraction, sugars, organic acids, and essential oil was significantly higher in etherel treated leaves than in the control. However, [U-14C]-sucrose feeding led to decreased label incorporation in the ethanol-soluble fraction, sugars, organic acids, and essential oils compared to the control. When 14CO2 was fed to GA treated leaves, label incorporation in ethanol-insoluble fraction, sugars, and oils was significantly higher than in the control. In contrast, when [U-14C]-sucrose was fed the incorporation in the ethanol soluble fraction, sugars, organic acids, and oil was significantly lower than in the control. Hence the hormone treatment induces a differential utilization of precursors for oil biosynthesis and accumulation and differences in partitioning of label between leaf and stem. Etherel and GA influence the partitioning of primary photosynthetic metabolites and thus modify plant growth and essential oil accumulation.

Probing DMPG vesicle surface with a cationic aqueous soluble spin label

A small, highly aqueous soluble, deuterated, cationic spin label, 4-trimethylammonium-2,2,6,6-tetramethylpiperidine-d 17-1-oxyl iodide (dCAT1), was used to directly monitor the negatively charged DMPG vesicle surface in order to test a recent suggestion (Riske et al., Chem. Phys. Lipids, 89 (1997) 31–44) that alterations in the surface potential accompanied apparent phase transitions observed by light scattering. The temperature dependence of the label partition between the lipid surface and the aqueous medium indicated an increase in the surface potential at the gel to liquid–crystal transition, supporting the previous suggestion. Results at the phase transition occurring at a higher temperature were less definitive. Although some change in the dCAT1 ESR spectra was observed, the interpretation of the phenomena is still rather unclear. DMPG surface potentials were estimated from the dCAT1 partition ratios (surface label moles/total label moles), using a simple two-sites model, where the electrostatic potential is zero everywhere but at the vesicle surface, and the interaction between the spin label and the membrane surface is chiefly electrostatic. The Gouy–Chapman–Stern model predicts surface potentials similar to those observed, although the measured decrease in the surface potential with ionic strength is somewhat steeper than that predicted by the model.

In vitro quantitative determination of phospholipid adducts of chloroform intermediates in hepatic and renal microsomes from different rodent strains

We have comparatively studied in vitro the oxidative and reductive pathways of chloroform metabolism in hepatic and renal microsomes of rodent strains used for carcinogenicity testing (B6C3F1 mice, Osborne Mendel and Sprague Dawley rats). To this aim we exploited the regioselective binding of phosgene to phospholipid (PL) polar heads and of dichloromethyl radical to PL fatty acyl chains, using a method based on the chemical transmethylation of PL adducts, followed by phase partitioning of the resulting products (De Biasi et al., 1992). The analysis of results let us to conclude at first that a 14C label partitioning by 89.2 (±6.5)% or 13.7 (±5.0)% in the aqueous phase is typical of the PL adduct with phosgene (PL-PHOS) or with dichloromethyl radical (PL-RAD), respectively. Metabolism of 0.1 mM CHCl 3 was mainly oxidative in all the samples, being hepatic microsomes more active than renal ones by about one order of magnitude and levels of CHCl 3-derived PL adducts in B6C3F1 mouse liver microsomes higher than in rat samples. At 5 mM CHCl 3, total levels of PL adducts in renal microsomes reached levels almost similar to those found in liver microsomes. However, while B6C3F1 mouse kidney microsomes produced both reactive metabolites, similarly as the hepatic samples, Osborne Mendel rat kidney microsomes bioactivated CHCl 3 only reductiveiy, producing the radical. The relevance of this finding depends on the fact that phosgene is known to be the major cause of CHCl 3 toxicity, based on data with the rat liver and mouse liver and kidney, while nephrotoxicity in rats occurs with minimal production of COCl 2. Chloroform reductive bioactivation may therefore provide a reasonable explanation for the toxicity of chloroform to the rat kidney. The same finding may be of interest in elucidating the metabolic reasons of the chloroform-induced kidney tumors in Osborne Mendel rats.

Lactate distribution and metabolism during and after anoxia in the turtle, Chrysemys picta bellii.

To determine the fate of lactate during and after prolonged anoxia, 14C-labeled lactate was injected into turtles after 2 h of a 6-h submergence at 20 degrees C. 14C activities of plasma and chamber water were tested at intervals during anoxia and also in expired air during 39 h of recovery. Partitioning of label in major body compartments [extracellular fluid (ECF), intracellular fluid (ICF), and shell] and 14C activity and glycogen in selected tissues (heart, liver, and muscle) were measured after anoxia (n = 7) and after recovery (n = 6). Shell 14C and [lactate] were extensively measured on six anoxic turtles. During anoxia all 14C remained in the animal indicating no urine production. At 6 h of anoxia 47% of recovered 14C, presumably still as lactate, was in the ECF, 27% in the ICF, and 30% in the shell. During recovery, plasma [lactate] fell from 35 to 5 meq, but surrounding water and expired air accounted for only 9 and 8%, respectively, of recovered label. The ICF portion grew to 41%, associated with a recovery in tissue glycogen. The shell still had 22% of total label. We conclude that, during recovery from anoxia, lactate is predominantly resynthesized to glycogen, and only a small fraction is directly oxidized. During anoxia, however, lactate is widely distributed in the body, and a surprisingly large and functionally significant fraction resides in the shell.

Thymidine incorporation of bacteria sequentially extracted from soil using repeated homogenization-centrifugation

Bacteria were sequentially extracted from soil into a water suspension after shaking soil with water or mixing it in a blender followed by a low-speed centrifugation. Bacteria, which were released only after several cycles of homogenization-centrifugation, had higher growth rates as judged from thymidine and leucine incorporation, whereas bacteria that were more readily released by a gentle shaking procedure had the lowest growth rate. This indicated that bacteria more tightly bound to soil particles were growing faster than those that were more easily released into the water suspension. The same pattern was found both in an agricultural and a forest soil, with contrasting pH and organic matter content, and irrespective of whether the bacteria were labeled before or after the centrifugation steps. The different growth rates of the bacteria could not be explained by different partitioning of label between different macromolecules, different cell size, different viability of the bacteria, or different dilution of the added radioactive substrate in the different homogenization-centrifugation fractions. The total amount of phospholipid fatty acids per bacterial cell was also similar in the different fractions. Different composition of the bacterial communities in the different homogenization-centrifugation fractions was indicated by a gradually altered phospholipid fatty acid pattern of the extracted bacteria, and an increased hydrophobicity of the bacteria released only after several homogenization-centrifugation treatments.

P-C1-10 Fluorescence label partition in doped phospolipid monolayers: Popp K, Diehl HA Inst. of Exp. Phys., Univ. of Bremen (FRG)

P-C1-10 Fluorescence label partition in doped phospolipid monolayers: Popp K, Diehl HA Inst. of Exp. Phys., Univ. of Bremen (FRG)

Transfer of phospholipase A-resistant pyrene-dialkyl-glycerophosphocholine to plasma lipoproteins: differences between Lp[a] and LDL.

1-O-Hexadecyl-2-O-pyrenedecanyl-sn-glycero-3-phosphocholine, a non-hydrolyzable fluorescent diether analog of phosphatidylcholine (PC), was synthesized as a probe for studying phospholipid transfer to different lipoprotein classes with potential phospholipase activities. After incubation of total human plasma with the new probe at 37 degrees C for 4.5 h, a characteristic partition between the main lipoprotein fractions was observed. The fluorescent lipid was not degraded under these conditions and, therefore, served as a measure for choline glycerophospholipid distribution between plasma lipoproteins. In low density lipoprotein (LDL) and high density lipoprotein-3 (HDL3) the fluorescent PC analog showed only monomer fluorescence, whereas in Lp[a] and HDL2 monomer and excimer fluorescence were observed, indicating that the fluorescent phosphatidylcholine analog was incorporated into the respective lipoproteins to a different extent. According to the increased pyrene excimer fluorescence in Lp[a] compared with LDL the labeled phosphatidylcholine must be enriched and/or clustered in Lp[a]. Data from phospholipid and total fluorescence analyses are compatible with the assumption of higher label concentration in Lp[a]. On the other hand, transfer rates for serum protein-catalyzed lipid transport into isolated Lp[a] were slower as compared to LDL. It is suggested that slower lipid transfer to Lp[a] under these conditions is due to the decreased lipid mobility in the Lp[a] surface, whereas the higher extent of label partition into Lp[a] as observed in total plasma might be due to the higher affinity of apolipoproteins for phosphatidylcholine in Lp[a] (Sommer, A., et al. 1992. J. Biol. Chem. 267: 24217-24222). The use of a fluorescent dialkyl- instead of diacyl-glycerophosphocholine for transfer studies was mandatory, as we found that lipoproteins contained phospholipase A2 activity toward long-chain phosphatidylcholine. The lipoprotein-associated phospholipase A2 was three times more active in Lp[a] than in LDL. The degradation products formed by the phospholipase, fatty acids, and lyso-PC may add to the high atherogenic potential of Lp[a].

Stability of protein formulations: investigation of surfactant effects by a novel EPR spectroscopic technique.

Surfactants are known to stabilize proteins and are often employed as additives in protein formulations. We have developed a method to study the interaction of these formulation additives with proteins by using the partitioning behavior of a spin label. In protein-free formulations, 16-doxyl stearic acid partitions into micelles above the critical micelle concentration (CMC) of the surfactant and gives rise to composite electron paramagnetic resonance (EPR) spectra composed of spectra from "free" label and "rotationally hindered" label. We compute the fraction of micelle-associated label by factor analysis and generate a label partition curve. When protein is added to the formulation, surfactant-protein aggregates form at concentrations below the surfactant's CMC. Partitioning of the label into these aggregates causes the EPR spectrum to reflect hindered rotation of the label at lower surfactant concentrations than in the protein-free solutions. A simple model of label partitioning shows that these partitioning shifts can be correlated to the surfactant:protein binding stoichiometry. We have studied the interactions of various non-ionic surfactants like Brij and Tween with recombinant human growth hormone and recombinant human interferon-gamma and obtained corresponding binding stoichiometries. These binding stoichiometries match those obtained by other techniques. This technique offers a new method for estimating the protein:surfactant binding stoichiometries.

Effect of high electrolyte concentration on the cooperativity of the main phase-transition of DPPC

we have investigated the effect on the cooperativity of the Pp. - L~ main phase transition of multilamellae of dipalmitoylphosphatidylcholine IDPPC) of a I : I electrolyte up to the concentration of 3 M in the dispersion medium. By using the ESR spectroscopy with the spin label partition technique, the mean size of the cooperative unit, v~ at the centre of the transition has been calculated according to the Zimm and Bragg (J. Chem Phys. 31 (1959) 526-535) theory of cooperative transitions. By increasing the electrolyte concentration in the dispersion medium the v~ value increases suggesting that the lipid molecules in the bilayers become more tigthly packed. Screening of the electrostatic interactions between the dipoles on the polar head of the DPPC molecules with increase of the hydrophobic interactions as well as modification of the structure of the water layer around the phosphocholine fragment of DPPCS can explain the results obtained.