The human factors of computer input using a mouse device

This thesis investigates how people select items from a computer display using the mouse input device. The term computer mouse refers to a class of input devices which share certain features, but these may have different characteristics which influence the ways in which people use the device. Although task completion time is one of the most commonly used performance measures for input device evaluation, there is no consensus as to its definition. Furthermore most mouse studies fail to provide adequate assurances regarding its correct measurement.Therefore precise and accurate timing software were developed which permitted the recording of movement data which by means of automated analysis yielded the device movements made. Input system gain, an important task parameter, has been poorly defined and misconceptualized in most previous studies. The issue of gain has been clarified and investigated within this thesis. Movement characteristics varied between users and within users, even for the same task conditions. The variables of target size, movement amplitude, and experience exerted significant effects on performance. Subjects consistently undershot the target area. This may be a consequence of the particular task demands. Although task completion times indicated that mouse performance had stabilized after 132 trials the movement traces, even of very experienced users, indicated that there was still considerable room for improvement in performance, as indicated by the proportion of poorly made movements. The mouse input device was suitable for older novice device users, but they took longer to complete the experimental trials. Given the diversity and inconsistency of device movements, even for the same task conditions, caution is urged when interpreting averaged grouped data. Performance was found to be sensitive to; task conditions, device implementations, and experience in ways which are problematic for the theoretical descriptions of device movement, and limit the generalizability of such findings within this thesis.

In vivo and in vitro aspects of imidazoline-2(1/2)sites within the central nervous system

The in vivo and in vitro characteristics of the I2 binding site were probed using the technique of drug discrimination and receptor autoradiography. Data presented in this thesis indicates the I2 ligand 2-BFI generates a cue in drug discrimination. Further studies indicated agmatine, a proposed endogenous imidazoline ligand, and a number of imidazoline and imidazole analogues of 2-BFI substitute significantly for 2-BFI. In addition to specific I2 ligands the administration of NRl's (noradrenaline reuptake inhibitors), the sympathomimetic d-amphetamine, the α1-adrenoceptor agonist methoxamine, but not the β1 agonist dobutamine or the β2 agonist salbutamol, gave rise to significant levels of substitution for the 2-BFI cue. The administration of the α1-adrenoceptor antagonist WB4101, prior to 2- BFI itself significantly reduced levels of 2-BFI appropriate responding. Administration of the reversible MAO-A inhibitors moclobemide and Ro41-1049, but not the reversible MAO-B inhibitors lazabemide and Ro16-6491, gave rise to potent dose dependent levels of substitution for the 2-BFI cue. Further studies indicated the administration of a number of β-carbolines and the structurally related indole alkaloid ibogaine also gave rise to dose dependent significant levels of substitution. Due to the relationship of indole alkaloids to serotonin the 5-HT releaser fenfluramine and a number of SSRI's (selective serotonin reuptake inhibitor) were also administered and these compounds gave rise to significant partial (20-80% responses to the 2-BFI lever) levels of substitution. The autoradiographical studies reported here indicate [3H]2-BFI labels I2 sites within the rat arcuate nucleus, area postrema, pineal gland, interpeduncular nucleus and subfornical organ. Subsequent experiments confirmed that the drug discrimination dosing schedule significantly increases levels of [3H]2-BFI 12 binding within two of these nuclei. However, levels of [3H]2-BFI specific binding were significantly reduced within four of these nuclei after chronic treatment with the irreversible MAO inhibitors deprenyl and tranylcypromine but not pargyline, which only reduced levels significantly in two. Further autoradiographical studies indicated that the distribution of [3H]2-BFI within the C57/B mouse compares favourably to that within the rat. Comparison of these levels of binding to those from transgenic mice who over-express MAO-B indicates two possibly distinct populations of [3H]2-BFI 12 sites exist in mouse brain. The data presented here indicates the 2-BFI cue is associated with the selective activation of α1-adrenoceptors and possibly 5-HT receptors. 2-BFI trained rats recognise reversible MAO-A but not MAO-B inhibitors. However, data within this thesis indicates the autoradiographical distribution of I2 sites bears a closer resemblance to that of MAO-B not MAO-A and further studies using transgenic mice that over-express MAO-B suggests a non-MAO-B I2 site exists in mouse brain.

Phospholipid chlorohydrin induces leukocyte adhesion to Apoe(-/-) mouse arteries via upregulation of P-selectin

HOCl-modified low-density lipoprotein (LDL) has proinflammatory effects, including induction of inflammatory cytokine production, leukocyte adhesion, and ROS generation, but the components responsible for these effects are not completely understood. HOCl and the myeloperoxidase-H2O2-halide system can modify both protein and lipid moieties of LDL and react with unsaturated phospholipids to form chlorohydrins. We investigated the proinflammatory effects of 1-stearoyl-2-oleoyl-sn-3-glycerophosphocholine (SOPC) chlorohydrin on artery segments and spleen-derived leukocytes from ApoE-/- and C57 Bl/6 mice. Treatment of ApoE-/- artery segments with SOPC chlorohydrin, but not unmodified SOPC, caused increased leukocyte-arterial adhesion in a time- and concentration-dependent manner. This could be prevented by pretreatment of the artery with P-selectin or ICAM-1-blocking antibodies, but not anti-VCAM-1 antibody, and immunohistochemistry showed that P-selectin expression was upregulated. However, chlorohydrin treatment of leukocytes did not increase expression of adhesion molecules LFA-1 or PSGL-1, but caused increased release of ROS from PMA-stimulated leukocytes by a CD36-dependent mechanism. The SOPC chlorohydrin-induced adhesion and ROS generation could be abrogated by pretreatment of the ApoE-/- mice with pravastatin or a nitrated derivative, NCX 6550. These findings suggest that phospholipid chlorohydrins formed in HOCl-treated LDL could contribute to the proinflammatory effects observed for this modified lipoprotein in vitro.

Studies on the antiobesity effect of zinc-α2-glycoprotein in the ob/ob mouse

OBJECTIVE: To investigate the mechanism of the lipid depletion by zinc-a(2)-glycoprotein (ZAG). DESIGN: Studies were conducted in the ob/ob mouse, or on isolated adipocytes from these animals or their lean counterparts. RESULTS: Treatment of these animals for 15 days with ZAG (100? µg, intravenously, daily) resulted in a reduction of body weight of 6.55? g compared with phosphate-buffered saline-treated controls, without a change in food or water intake, but with a 0.4?°C rise in rectal temperature. ZAG-treated mice had a 30% reduction in carcass fat mass and a twofold increase in weight of brown adipose tissue. Epididymal adipocytes from ZAG-treated mice showed an increased expression of ZAG and hormone-sensitive lipase (HSL), and this was maintained for a further 3 days in the absence of ZAG. There was an increased lipolytic response to isoproterenol, which was retained for 3 days in vitro in the absence of ZAG. Expression of HSL was also increased in subcutaneous and visceral adipose tissue, as was also adipose triglyceride lipase (ATGL). There was a rapid loss of labelled lipid from epididymal adipose tissue of ZAG-treated mice, but not from the other depots, reflecting the difference in sensitivity to lipolytic stimuli. The increased expression of HSL and ATGL may involve the extracellular signal-regulated kinase (ERK) pathway, as the active (phospho) form was upregulated in all adipose depots after ZAG administration, whereas in vitro studies showed induction of HSL and ATGL by ZAG to be attenuated by PD98059, an inhibitor of the ERK pathway. CONCLUSION: These results suggest that ZAG not only induces direct lipolysis, but also sensitizes adipose tissue to other lipolytic stimuli.

Visfatin regulates insulin secretion, insulin receptor signalling and mRNA expression of diabetes-related genes in mouse pancreatic beta-cells

The role of the adipocyte-derived factor visfatin in metabolism remains controversial, although some pancreatic ß-cell-specific effects have been reported. This study investigated the effects of visfatin upon insulin secretion, insulin receptor activation and mRNA expression of key diabetes-related genes in clonal mouse pancreatic ß-cells. ß-TC6 cells were cultured in RPMI 1640 and were subsequently treated with recombinant visfatin. One-hour static insulin secretion was measured by ELISA. Phospho-specific ELISA and western blotting were used to detect insulin receptor activation. Real-time SYBR Green PCR array technology was used to measure the expression of 84 diabetes-related genes in both treatment and control cells. Incubation with visfatin caused significant changes in the mRNA expression of several key diabetes-related genes, including marked up-regulation of insulin (9-fold increase), hepatocyte nuclear factor (HNF)1ß (32-fold increase), HNF4a (16-fold increase) and nuclear factor ?B (40-fold increase). Significant down-regulation was seen in angiotensin-converting enzyme (-3.73-fold) and UCP2 (-1.3-fold). Visfatin also caused a significant 46% increase in insulin secretion compared to control (P<0.003) at low glucose, and this increase was blocked by co-incubation with the specific nicotinamide phosphoribosyltransferase inhibitor FK866. Both visfatin and nicotinamide mononucleotide induced activation of both insulin receptor and extracellular signal-regulated kinase (ERK)1/2, with visfatin-induced insulin receptor/ERK1/2 activation being inhibited by FK866. We conclude that visfatin can significantly regulate insulin secretion, insulin receptor phosphorylation and intracellular signalling and the expression of a number of ß-cell function-associated genes in mouse ß-cells.

α7 nicotinic receptor-mediated astrocytic gliotransmitter release:Aβ effects in a preclinical Alzheimer's mouse model

It is now recognized that astrocytes participate in synaptic communication through intimate interactions with neurons. A principal mechanism is through the release of gliotransmitters (GTs) such as ATP, D-serine and most notably, glutamate, in response to astrocytic calcium elevations. We and others have shown that amyloid-β (Aβ), the toxic trigger for Alzheimer's disease (AD), interacts with hippocampal α7 nicotinic acetylcholine receptors (nAChRs). Since α7nAChRs are highly permeable to calcium and are expressed on hippocampal astrocytes, we investigated whether Aβ could activate astrocytic α7nAChRs in hippocampal slices and induce GT glutamate release. We found that biologically-relevant concentrations of Aβ1-42 elicited α7nAChR-dependent calcium elevations in hippocampal CA1 astrocytes and induced NMDAR-mediated slow inward currents (SICs) in CA1 neurons. In the Tg2576 AD mouse model for Aβ over-production and accumulation, we found that spontaneous astrocytic calcium elevations were of higher frequency compared to wildtype (WT). The frequency and kinetic parameters of AD mice SICs indicated enhanced gliotransmission, possibly due to increased endogenous Aβ observed in this model. Activation of α7nAChRs on WT astrocytes increased spontaneous inward currents on pyramidal neurons while α7nAChRs on astrocytes of AD mice were abrogated. These findings suggest that, at an age that far precedes the emergence of cognitive deficits and plaque deposition, this mouse model for AD-like amyloidosis exhibits augmented astrocytic activity and glutamate GT release suggesting possible repercussions for preclinical AD hippocampal neural networks that contribute to subsequent cognitive decline. © 2013 Pirttimaki et al.

The prevalence and phenotype of activated microglia/macrophages within the spinal cord of the hyperostotic mouse (twy/twy) changes in response to chronic progressive spinalcord compression:implications for human cervical compressive myelopathy

Background:Cervical compressive myelopathy, e.g. due to spondylosis or ossification of the posterior longitudinal ligament is a common cause of spinal cord dysfunction. Although human pathological studies have reported neuronal loss and demyelination in the chronically compressed spinal cord, little is known about the mechanisms involved. In particular, the neuroinflammatory processes that are thought to underlie the condition are poorly understood. The present study assessed the localized prevalence of activated M1 and M2 microglia/macrophages in twy/twy mice that develop spontaneous cervical spinal cord compression, as a model of human disease.Methods:Inflammatory cells and cytokines were assessed in compressed lesions of the spinal cords in 12-, 18- and 24-weeks old twy/twy mice by immunohistochemical, immunoblot and flow cytometric analysis. Computed tomography and standard histology confirmed a progressive spinal cord compression through the spontaneously development of an impinging calcified mass.Results:The prevalence of CD11b-positive cells, in the compressed spinal cord increased over time with a concurrent decrease in neurons. The CD11b-positive cell population was initially formed of arginase-1- and CD206-positive M2 microglia/macrophages, which later shifted towards iNOS- and CD16/32-positive M1 microglia/macrophages. There was a transient increase in levels of T helper 2 (Th2) cytokines at 18 weeks, whereas levels of Th1 cytokines as well as brain-derived neurotrophic factor (BDNF), nerve growth factor (NGF) and macrophage antigen (Mac) -2 progressively increased.Conclusions:Spinal cord compression was associated with a temporal M2 microglia/macrophage response, which may act as a possible repair or neuroprotective mechanism. However, the persistence of the neural insult also associated with persistent expression of Th1 cytokines and increased prevalence of activated M1 microglia/macrophages, which may lead to neuronal loss and demyelination despite the presence of neurotrophic factors. This understanding of the aetiopathology of chronic spinal cord compression is of importance in the development of new treatment targets in human disease. © 2013 Hirai et al.

Effect of maternal protein restriction during pregnancy and postweaning high-fat feeding on diet-induced thermogenesis in adult mouse offspring

Purpose: Prenatal undernutrition followed by postweaning feeding of a high-fat diet results in obesity in the adult offspring. In this study, we investigated whether diet-induced thermogenesis is altered as a result of such nutritional mismatch. Methods: Female MF-1 mice were fed a normal protein (NP, 18 % casein) or a protein-restricted (PR, 9 % casein) diet throughout pregnancy and lactation. After weaning, male offspring of both groups were fed either a high-fat diet (HF; 45 % kcal fat) or standard chow (C, 7 % kcal fat) to generate the NP/C, NP/HF, PR/C and PR/HF adult offspring groups (n = 7-11 per group). Results: PR/C and NP/C offspring have similar body weights at 30 weeks of age. Postweaning HF feeding resulted in significantly heavier NP/HF offspring (P <0.01), but not in PR/HF offspring, compared with their chow-fed counterparts. However, the PR/HF offspring exhibited greater adiposity (P <0.01) v the NP/HF group. The NP/HF offspring had increased energy expenditure and increased mRNA expression of uncoupling protein-1 and β-3 adrenergic receptor in the interscapular brown adipose tissue (iBAT) compared with the NP/C mice (both at P <0.01). No such differences in energy expenditure and iBAT gene expression were observed between the PR/HF and PR/C offspring. Conclusions: These data suggest that a mismatch between maternal diet during pregnancy and lactation, and the postweaning diet of the offspring, can attenuate diet-induced thermogenesis in the iBAT, resulting in the development of obesity in adulthood. © 2014 Springer-Verlag Berlin Heidelberg.

Metabolic induction and early responses of mouse blastocyst developmental programming following maternal low protein diet affecting life-long health

Previously, we have shown that a maternal low protein diet, fed exclusively during the preimplantation period of mouse development (Emb-LPD), is sufficient to induce by the blastocyst stage a compensatory growth phenotype in late gestation and postnatally, correlating with increased risk of adult onset cardiovascular disease and behavioural dysfunction. Here, we examine mechanisms of induction of maternal Emb-LPD programming and early compensatory responses by the embryo. Emb-LPD induced changes in maternal serum metabolites at the time of blastocyst formation (E3.5), notably reduced insulin and increased glucose, together with reduced levels of free amino acids (AAs) including branched chain AAs leucine, isoleucine and valine. Emb-LPD also caused reduction in the branched chain AAs within uterine fluid at the blastocyst stage. These maternal changes coincided with an altered content of blastocyst AAs and reduced mTORC1 signalling within blastocysts evident in reduced phosphorylation of effector S6 ribosomal protein and its ratio to total S6 protein but no change in effector 4E-BP1 phosphorylated and total pools. These changes were accompanied by increased proliferation of blastocyst trophectoderm and total cells and subsequent increased spreading of trophoblast cells in blastocyst outgrowths. We propose that induction of metabolic programming following Emb-LPD is achieved through mTORC1signalling which acts as a sensor for preimplantation embryos to detect maternal nutrient levels via branched chain AAs and/or insulin availability. Moreover, this induction step associates with changes in extra-embryonic trophectoderm behaviour occurring as early compensatory responses leading to later nutrient recovery. © 2012 Fleming et al.

Maternal low-protein diet during mouse pre-implantation development induces vascular dysfunction and altered renin-angiotensin-system homeostasis in the offspring

Environmental perturbations during early mammalian development can affect aspects of offspring growth and cardiovascular health. We have demonstrated previously that maternal gestational dietary protein restriction in mice significantly elevated adult offspring systolic blood pressure. Therefore, the present study investigates the key mechanisms of blood pressure regulation in these mice. Following mating, female MF-1 mice were assigned to either a normal-protein diet (NPD; 18% casein) or an isocaloric low-protein diet throughout gestation (LPD; 9% casein), or fed the LPD exclusively during the pre-implantation period (3.5d) before returning to the NPD for the remainder of gestation (Emb-LPD). All offspring received standard chow. At 22 weeks, isolated mesenteric arteries from LPD and Emb-LPD males displayed significantly attenuated vasodilatation to isoprenaline (P=0.04 and P=0.025, respectively), when compared with NPD arteries. At 28 weeks, stereological analysis of glomerular number in female left kidneys revealed no significant difference between the groups. Real-time RT-PCR analysis of type 1a angiotensin II receptor, Na /K ATPase transporter subunits and glucocorticoid receptor expression in male and female left kidneys revealed no significant differences between the groups. LPD females displayed elevated serum angiotensin-converting enzyme (ACE) activity (P=0.044), whilst Emb-LPD males had elevated lung ACE activity (P=0.001), when compared with NPD offspring. These data demonstrate that elevated offspring systolic blood pressure following maternal gestational protein undernutrition is associated with impaired arterial vasodilatation in male offspring, elevated serum and lung ACE activity in female and male offspring, respectively, but kidney glomerular number in females and kidney gene expression in male and female offspring appear unaffected. © 2010 The Authors.

Low protein diet fed exclusively during mouse oocyte maturation leads to behavioural and cardiovascular abnormalities in offspring

Early embryonic development is known to be susceptible to maternal undernutrition, leading to a disease-related postnatal phenotype. To determine whether this sensitivity extended into oocyte development, we examined the effect of maternal normal protein diet (18% casein; NPD) or isocaloric low protein diet (9% casein; LPD) restricted to one ovulatory cycle (3.5 days) prior to natural mating in female MF-1 mice. After mating, all females received NPD for the remainder of gestation and all offspring were litter size adjusted and fed standard chow. No difference in gestation length, litter size, sex ratio or postnatal growth was observed between treatments. Maternal LPD did, however, induce abnormal anxiety-related behaviour in open field activities in male and female offspring (P <0.05). Maternal LPD offspring also exhibited elevated systolic blood pressure (SBP) in males at 9 and 15 weeks and in both sexes at 21 weeks (P <0.05). Male LPD offspring hypertension was accompanied by attenuated arterial responsiveness in vitro to vasodilators acetylcholine and isoprenaline (P <0.05). LPD female offspring adult kidneys were also smaller, but had increased nephron numbers (P <0.05). Moreover, the relationship between SBP and kidney or heart size or nephron number was altered by diet treatment (P <0.05). These data demonstrate the sensitivity of mouse maturing oocytes in vivo to maternal protein undernutrition and identify both behavioural and cardiovascular postnatal outcomes, indicative of adult disease. These outcomes probably derive from a direct effect of protein restriction, although indirect stress mechanisms may also be contributory. Similar and distinct postnatal outcomes were observed here compared with maternal LPD treatment during post-fertilization preimplantation development which may reflect the relative contribution of the paternal genome. © Journal compilation © 2008 The Physiological Society.

Role of transglutaminase 1 in stabilisation of intercellular junctions of the vascular endothelium

Microvascular endothelial monolayers from mouse myocardium (MyEnd) cultured for up to 5 days postconfluency became increasingly resistant to various barrier-compromising stimuli such as low extracellular Ca2+ and treatment with the Ca2+ ionophore A23187 and with the actin depolymerising compound cytochalasin D. In contrast, microvascular endothelial monolayers from mouse lung microvessels (PulmEnd) remained sensitive to these conditions during the entire culture period which corresponds to the well-known in vivo sensitivity of the lung microvasculature to Ca2+depletion and cytochalasin D treatment. One molecular difference between pulmonary and myocardial endothelial cells was found to be transglutaminase 1 (TGase1) which is strongly expressed in myocardial endothelial cells but is absent from pulmonary endothelial cells. Resistance of MyEnd cells to barrier-breaking conditions correlated strongly with translocation of TGase1 to intercellular junctions. Simultaneous inhibition of intracellular and extracellular TGase activity by monodansylcadaverine (MDC) strongly weakened barrier properties of MyEnd monolayers, whereas inhibition of extracellular TGases by the membrane-impermeable active site-directed TGase inhibitor R281 did not reduce barrier properties. Weakening of barrier properties could be also induced in MyEnd cells by downregulation of TGase1 expression using RNAi-based gene silencing. These findings suggest that crosslinking activity of intracellular TGase1 at intercellular junctions may play a role in controlling barrier properties of endothelial monolayers.

Influence of Egr-1 in cardiac tissue-derived mesenchymal stem cells in response to glucose variations

Mesenchymal stem cells (MSCs) represent a promising cell population for cell therapy and regenerative medicine applications. However, how variations in glucose are perceived by MSC pool is still unclear. Since, glucose metabolism is cell type and tissue dependent, this must be considered when MSCs are derived from alternative sources such as the heart. The zinc finger transcription factor Egr-1 is an important early response gene, likely to play a key role in the glucose-induced response. Our aim was to investigate how short-term changes in in vitro glucose concentrations affect multipotent cardiac tissue-derived MSCs (cMSCs) in a mouse model of Egr-1 KO (Egr-1-/-). Results showed that loss of Egr-1 does not significantly influence cMSC proliferation. In contrast, responses to glucose variations were observed in wt but not in Egr-1 -/- cMSCs by clonogenic assay. Phenotype analysis by RT-PCR showed that cMSCs Egr-1-/- lost the ability to regulate the glucose transporters GLUT-1 and GLUT-4 and, as expected, the Egr-1 target genes VEGF, TGFβ-1, and p300. Acetylated protein levels of H3 histone were impaired in Egr-1-/- compared to wt cMSCs. We propose that Egr-1 acts as immediate glucose biological sensor in cMSCs after a short period of stimuli, likely inducing epigenetic modifications. © 2014 Daniela Bastianelli et al.

Nucleoside diphosphate kinase--a component of the [Na+1]- and [Cl-]-sensitive phosphorylation cascade in human and murine airway epithelium

We have shown that proteins within apically enriched fractions of human nasal respiratory epithelium vary their phosphohistidine content with ambient [Cl-] and other anion concentrations. This membrane-delimited phosphorylation cascade includes a multifunctional protein histidine kinase - nucleoside diphosphate kinase (NDPK). NDPK is itself a cascade component in both human and ovine airway, the self-phosphorylation of which is inhibited selectively by [Na+] in the presence of ATP (but not GTP). These findings led us to propose the existence of a dual anion-/cation-controlled phosphorylation-based "sensor" bound to the apical membrane. The present study showed that this cascade uses ATP to phosphorylate a group of proteins above 45 kDa (p45-group, identities unknown). Additionally, the Cl- dependence of ATP (but not GTP) phosphorylation is conditional on phosphatase activity and that interactions exist between the ATP- and GTP-phosphorylated components of the cascade under Cl--free conditions. As a prelude to studies in cystic fibrosis (CF) mice, we showed in the present study that NDPK is present and functionally active in normal murine airway. Since NDPK is essential for UTP synthesis and regulates fetal gut development, G proteins, K+channels, neutrophil-mediated inflammation and pancreatic secretion, the presence of ion-regulated NDPK protein in mouse airway epithelium might aid understanding of the pathogenesis of CF.

Toward prediction of class II mouse major histocompatibility complex peptide binding affinity:in silico bioinformatic evaluation using partial least squares, a robust multivariate statistical technique

The accurate identification of T-cell epitopes remains a principal goal of bioinformatics within immunology. As the immunogenicity of peptide epitopes is dependent on their binding to major histocompatibility complex (MHC) molecules, the prediction of binding affinity is a prerequisite to the reliable prediction of epitopes. The iterative self-consistent (ISC) partial-least-squares (PLS)-based additive method is a recently developed bioinformatic approach for predicting class II peptide−MHC binding affinity. The ISC−PLS method overcomes many of the conceptual difficulties inherent in the prediction of class II peptide−MHC affinity, such as the binding of a mixed population of peptide lengths due to the open-ended class II binding site. The method has applications in both the accurate prediction of class II epitopes and the manipulation of affinity for heteroclitic and competitor peptides. The method is applied here to six class II mouse alleles (I-Ab, I-Ad, I-Ak, I-As, I-Ed, and I-Ek) and included peptides up to 25 amino acids in length. A series of regression equations highlighting the quantitative contributions of individual amino acids at each peptide position was established. The initial model for each allele exhibited only moderate predictivity. Once the set of selected peptide subsequences had converged, the final models exhibited a satisfactory predictive power. Convergence was reached between the 4th and 17th iterations, and the leave-one-out cross-validation statistical terms - q2, SEP, and NC - ranged between 0.732 and 0.925, 0.418 and 0.816, and 1 and 6, respectively. The non-cross-validated statistical terms r2 and SEE ranged between 0.98 and 0.995 and 0.089 and 0.180, respectively. The peptides used in this study are available from the AntiJen database (http://www.jenner.ac.uk/AntiJen). The PLS method is available commercially in the SYBYL molecular modeling software package. The resulting models, which can be used for accurate T-cell epitope prediction, will be made freely available online (http://www.jenner.ac.uk/MHCPred).