Netlogo tm download windows 10

Sign Up Log In. Copy and paste the desired citation format or use the link below to download a file formatted for EndNote. All DeepDyve websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser. Open Advanced Search. DeepDyve requires Javascript to function. Please enable Javascript on your browser to continue. Ginovart, Marta The general aim is to promote the use of individual-based models biological agent-based models in teaching and learning contexts in life sciences and to make their progressive incorporation into academic curricula easier, complementing other existing modelling strategies more frequently used in the classroom.

Read Article. Download PDF. Share Full Text for Free. Web of Science. Let us know here. System error. Please try again! How was the reading experience on this article? The text was blurry Page doesn't load Other:. Details Include any more information that will help us locate the issue and fix it faster for you. Thank you for submitting a report! Submitting a report will send us an email through our customer support system.

Submit report Close. Recommended Articles Loading References Digital learning material for model building in molecular biology. Among patients in intensive care units ICUs , sepsis ranks as the second highest cause of mortality [ 6 ] and the 10th leading cause of death overall in the US [ 7 ].

An average of , sepsis cases occur annually, and this number continues to increase [ 6 ]. Sepsis in a hospitalized patient can lead to extended hospital stays and subsequently increase financial burdens.

Cross and Opal [ 10 ] discussed the lack of rapid, reliable assays available to identify the stage or severity of sepsis and to monitor the use of immunomodulatory therapy. Such assays are unavailable because of the complexity of the inflammatory response and the unpredictable nature of septic shock in individual patients; consequently increasing the difficulty of monitoring single or multiple components of inflammation with specific supportive therapies [ 10 , 11 ].

A significant past focus on modeling immune responses during sepsis has emerged in an effort to explore the complicated dynamic presentation of cells, tissues, and cytokines during infection. In , Kumar et al. In , Reynolds et al. Using a series of known and hypothesized kinetics of biological system components from the literature, mathematical models describe infectious disease processes by measuring steady states of various components in the immune system [ 14 ].

The agent-based model ABM , a powerful computational modeling technique, simulates complex nonlinear relationships between components and intuitively maps a realistic biological system by incorporating spatial effects and the stochastic nature of the immune response into model construction [ 16 , 17 ].

One key element of ABM includes agents, a collection of decision-making entities classified into types based on components described in the real-world system. Each type of agent executes behaviors that can mimic the system they represent when aggregated. Implementation of a predefined set of rules allows agents to move in a designed direction and arbitrarily interact with other agents in a spatial environment.

Agent behaviors are updated in various locations according to update rules executed at discrete time steps. ABM inherently captures repetitive spatial interactions between agents in a stochastic process or under a known probability distribution, making it a powerful tool to render valuable information and simulate a biological system.

Implementation of ABM requires well-established technology that relies on computers to explore dynamics beyond the reach of pure mathematical methods [ 18 , 19 ]. The inherent nature of computational structure allows ABM to be efficiently implemented on parallel computers [ 20 ].

An and his collaborators [ 21 — 23 ] developed a series of agent-based models to simulate behaviors of cells and cytokines in both the innate and adaptive immune system of a generalized inflammatory response. In , Wu et al. Recently, Dutta-Moscato et al.

In addition to modeling interactions between cells, Dong et al. Their approach explored hypothetical scenarios of AIR and potentially improved the understanding of molecular behaviors that could develop and expand to emergent behavior of the entire AIR system. In addition to these related work, there are studies that presented the application of agent-based models to simulate various types of disease progression [ 27 — 30 ].

Existing ABMs provide evidence that agent-based modeling is a valid approach for simulating disease progression [ 21 , 22 , 24 — 26 ]. By specifying the infected species, source of infection, and site of infection, the scope of the IMMABM allowed us to improve modeling approach accuracy without loss of generality. This IMMABM required that each interaction incorporated into the model was based on actual data from observations made during experimental infections in vivo or measurements made ex vivo or in vitro , thereby resulting in an incorporation of experimental data from publications related to mouse hepatic inflammation induced by Salmonella.

When data were not available, we extrapolated from related Gram-negative bacteria or other pathogens, keeping in mind that fidelity to actual Salmonella infections was necessary. Therefore, we summarized interactions among cells, tissues, and cytokines during mouse HIR and we calibrated quantitative changes in the HIR with experimental data and necessary mathematical expressions for agent modeling.

We attempted to calibrate variables based on unit relationships observed in the experimental systems. Simulated results from IMMABM showed that four distinct dynamic patterns emerge during mouse HIR: a healing response, persistent infection, a hyperinflammatory response, and organ dysfunction. Emerging simulations were verified through a pattern-oriented analysis found in available mouse experimental studies.

The liver, enriched with resident tissue macrophages Kupffer Cells , is recognized as a key organ of the immune system that is vital for elimination of a Salmonella infection [ 31 , 32 ]. Furthermore, immune responses to Salmonella infections have been investigated extensively [ 34 — 39 ].

Therefore, an abundance of data is available for accurate incorporation of relationships among variables agents in order to support our IMMABM. We used NetLogo 5.

The interface of NetLogo allows the modeler to set initial parameters and observe simulation results. We focused on the cellular interactions between liver sinusoid and hepatocytes in mouse. The interface of cellular interactions is comprised of five main compartments: liver sinusoid, sinusoid endothelial cells SECs , the space of Disse, the site of hepatocytes, and portal triad Fig 1A [ 42 ].

The Portal triad is a complex area including the hepatic artery, portal vein, and bile duct [ 42 ]. Blood flows from the portal triad area to the liver sinusoid, which carries blood-borne pathogens i. Salmonella to the site of hepatocytes. Hepatocytes are separated from the liver sinusoid by the space of Disse and sinusoid endothelial cells [ 42 ]. Kupffer Cells are distributed along sinusoid endothelial cells, and are able to ingest and kill the blood-borne Salmonella [ 43 ].

To mimic this liver structure, we divided the entire interface of NetLogo into five regions to represent the liver sinusoid, SECs, the space of Disse, the site of hepatocytes, and portal triad in the liver [ 43 ].

In the silico simulated environment, the probabilities that different agents cells, cytokines interact are more important than the actual physical morphology, which in vivo determines how these agents will interact. The choice of agents is directly comparable to the cell types and tissue organization formed in the liver. Therefore, the NetLogo setup is appropriate for this model. The initialized interface of NetLogo is shown in Fig 1B. Kupffer Cell numbers are approximately one-fourth the number of hepatocytes in the liver [ 42 ].

SEC numbers are approximately one-third the number of hepatocytes, and approximately one-eighth the number of mast cells exist compared to the number of hepatocytes [ 42 , 44 ]. For simulation size presented in this paper, the number of hepatocytes was initialized to 80, Considering the numeric proportion between hepatocytes, Kupffer Cells, SECs, and mast cells, we set the initial number of Kupffer Cells to 20,, SECs to 26,, and mast cells to 10, A Diagrams of 2D liver structure in mouse.

By assigning various values to the state variables, the agents were regulated to execute a series of functions based on various locations and environmental interfaces.

Interactions between agents were highly stochastic, and we incorporated mathematical expressions such as logistic growth functions, mass-action kinetics, Michaelis-Menten kinetics, and decay functions to quantitatively measure complicated biological processes. Those rules conformed to biological mechanisms of HIR. Data such as infiltration time of immune cells, replication rate of Salmonella , and degradation rate of associated mediators allowed us to advance the ABM application by mapping biological processes that occur during HIR to our IMMABM.

By integrating experimental data and mathematical expressions derived from hypothesized kinetics, we attempted to quantitatively simulate dynamic patterns of HIR. In our simulation, 1 tick representing 1 simulation step in the simulation software represented 1 hr in an actual biological process, and numeric counts of an agent were updated per tick to correspond to the biological response time in the experiments.

Incorporation of data from publications and our experience with Salmonella infections and infectious disease processes motivated us to select a total of 23 essential cells and cytokines as agent types in this IMMABM. In this paper, we use italic format to highlight agent type for convenience.

The rule system for these agents was based on the literature. A sequence of interactions among primary agents and primary agent behaviors during interactions are introduced in Section 2.

State variables associated with agent type were used to define various states of individual agents. Implementation of state variables is introduced in section 2. Experimental data showed that every neutrophil phagocytized approximately 3 to 13 Salmonella per hour, and every MDMI phagocytized approximately 1 to 7 Salmonella per hour [ 51 ].

In addition to death, Hepatocyte was also regenerated at a rate of 1. Kupffer Cells reside principally within the lumen of liver sinusoids, adherent to SECs that comprise blood vessel walls [ 43 ]. Kupffer Cells also released IL The apoptosis of Kupffer Cells occurs at a rate of 4.

MDMIs were activated from Monocytes between 6 hrs to 24 hrs post-infection [ 74 , 75 ]. The activation amount was calculated based on Michaelis-Menten kinetics, as discussed in Section 2.

B Cell released Antibody to form an Antibody - Salmonella complex, and the Antibody - Salmonella complex was killed by phagocytic cells, simulating opsonization [ 56 ]. The binding process is described in Section 2. An overview of agent behaviors is provided in S1 Table.

For example, Resting Neutrophils or Resting Monocytes moving to SECs were driven by Signals released from cytokines or cells [ 56 , 59 , 67 , 73 , 85 , 95 — 97 ]. Mass-action kinetics determined the number of moving Resting Neutrophil s or Resting Monocytes , as described in Section 2.

Released Salmonella randomly moved to the nearest Hepatocytes and damaged those Hepatocytes. For example, when an Antibody - Salmonella complex moved to one phagocytic cell, Antibody and Salmonella moved in the same direction for the same distance [ 55 ].

By assigning various values to the state variables, the agents were regulated to execute a series of functions based on locations and environment interfaces. During a simulation, some state variable are fixed through the simulation, and others change as the simulation runs.

The change in values of state variables is based on the change of agent behaviors during the simulation. A detailed description of agent behaviors in vivo is provided in S2 Table. If a state variable was equal to 1, individual agents that had that state variable would express specific attributes or execute biological functions. A comprehensive description of agent rule updates is presented in S1 Table. In order to calibrate quantitative changes in agent number during HIR, we used a standard logistic function to measure cell population increases, Michaelis-Menten kinetics to calibrate cytokine release, mass-action kinetics to calibrate the activation process of circulating neutrophils and monocytes, and a decay function to measure the natural process of apoptosis by cells or catabolism of cytokines.

For example, we calibrated the Salmonella population to increase using a standard logistic growth function [ 98 ] as follows:. Growth rates and carrying capacities of Salmonella varied when Salmonella replicated within various cells. Corresponding experimental data is presented in S3 Table.

We used a mass-action kinetics equation [ 99 ] to calibrate the activation process of circulating neutrophils as follows:. The release of cytokines obeyed trafficking machinery, and cytokines were released via protein-protein interactions initiated by ligand binding to receptors [ , ].

Therefore, we used Michaelis-Menten kinetics [ ] to calibrate the cytokine release process as follows:. In Eq 3 , C represents cytokine count and K max represents the maximum production rate of cytokines secretion by the cell. Cell denotes current numbers of the cell intending to release the cytokine and Cell half denotes cell numbers when half the maximum production rate of the cytokine was reached in the IMMABM.

Natural cell apoptosis or cytokine catabolism occurred at every tick; thus, we assumed that the decrease in cell or cytokine counts followed a simple decay function as follows:. In Eq 4 , C represents cell or cytokine count and Kc represents a constant decay rate for cells or cytokines. In addition to mathematical models, we calibrated experimental data such as replication rates of cells, production rate of cytokines, killing rates of Salmonella by phagocytic cells, activation rates of circulating neutrophils or monocytes, and apoptotic rate of cells or catabolism of cytokines from existing experimental studies.

We collected experimental data from studies that were most similar to our simulation setting. We also estimated parameters during simulation if data were not available from experimental studies. For example, we estimated that the CRP-opsonized debris moved to inflammatory cells e. An overview of estimated experimental data is provided in S3 Table. In order to smoothly translate estimated experimental data to agent-based modeling, we made corresponding assumptions in terms of data estimation.

In general, we assume the change in rate is constant because we observed changes in data of interests in most of experimental studies following linear curves. Some experimental data is comprised of multiple linear segments, and therefore we calibrated rates for each linear segment to measure various rates for multiple responding time periods. These multiple rates are explained in S3 Table. It was not possible to extrapolate the data for our agent-based model from one simple experimental model.

The strategy we used was to focus on mouse Salmonella infection studies that were published in papers available in the NCBI. When necessary, we used data from broader systems such as Gram-negative infections i.

Therefore, we are aware that some of these assumptions may not be correct. In IMMABM, we used agent count to represent cell number with the awareness that cytokine production rate has a unique experimental unit compared to cell number. Thereby, cytokine production rate had to be transformed into an agent number in order to make the experimental data consistent in IMMABM. Therefore, we used one agent count to represent one real experimental unit.

For example, we estimated that one phagocytic cell can bind 1. Therefore, we used one IL agent count to represent 1. Similarly, 1. Thus, we used one CRP agent count to represent 1. Data showed that 2. Unfortunately, however, NET structure is fragile, thereby making it difficult to quantify NETs in experiments [ ]. The rate at which NETs kill Salmonella was also difficult to establish [ ]. Data normality was checked using both histogram and quantile-quantile Q-Q plot. For normally distributed data, group comparisons were performed using one-way analysis of variance ANOVA.

For non-normally distributed data, Mann-Whitney U tests were conducted to compare groups. All tests were performed using R 3. The input data, converted as described to mathematical expressions and incorporated into the computer code, assembled cellular and molecular variables in order to generate a hypothetical immune response. Clinical and experimental data showed that the risk of patients dying from sepsis is significantly correlated to the initial dose of pathogen [ , ].

Therefore, we hypothesized that HIR would have a higher likelihood of progressing to septic shock and death if the infection was initially high.

To test this hypothesis, the IMMABM in silico simulations were carried out using Salmonella doses of counts, counts, counts, and counts, and replicated runs for each proposed Salmonella dose to explore the possible stochastic nature of the model.

Results from these simulations were initially generated to identify dynamic patterns of indicators in HIR with various initial Salmonella doses, as shown in Fig 2. The dose-response hypothesis test initially indicated that the HIR was correlated to Salmonella infection, which was consistent with experimental outcomes [ ].

We found four identifiable patterns in simulated HIR. Corresponding changes in the interface of NetLogo simulation were captured. Accumulation of Salmonella bacteria black areas and TNF-alpha yellow areas in situ. Note: 1 step is equivalent to 1 hr. When the initial infection with Salmonella was counts, the number of Hepatocyte Debris and CRP increased for the first 18 hrs of simulation but then progressively decreased to 0, demonstrating no additional pathology at later stages of the simulation.

We inferred that this combination of variables is similar to a host curing an infection, so we referred to it as a healing process Fig 3. We detected that a small number of hepatocytes less than 0. We also found that only a few neutrophils and monocytes less than cell counts were activated when the initial Salmonella infection was Ultimately, damaged hepatocytes were replaced with new healthy hepatocytes as the simulation proceeded Fig 4. Experimental studies in mice have shown early expression of pro-inflammatory cytokines in response to Salmonella infection [ ].

Similarly, the kinetics and amounts of secreted HMGB-1 correlated with the peak level of an HMGB-1 response seen in experimental observations if model size was taken into account [ ]. We observed that the increase in HMGB-1 levels began later in our model compared to production kinetics observed in in vitro stimulation assays [ ], However, kinetics of our model were consistent with the delayed contribution HMGB-1 is proposed to have during sepsis [ ].

Recruitment of monocytes to the liver rose sharply around 24 hrs after infection in our model, which is consistent to approximately 1 day in an actual experimental system [ ]. During actual infections, the decrease in bacterial load correlated with the influx of neutrophils [ ].

We observed a similar trend in the simulation Fig 4. We used CRP levels and Hepatocyte Debris to reflect the level of tissue damage that occurred after infection. A similar pattern of CRP concentrations was identified in healthy patients infected by bacteria in clinical cases [ ]. In some simulation replications, when the initial Salmonella infection was , the outcome more closely resembled a persistent infection, defined as the state in which Hepatocyte Debris , CRP , and Salmonella levels initially declined but subsequently increased to much higher levels before the infection was resolved at approximately 90 hrs.

Fig 5. Under this condition, Activated Neutrophil numbers declined along with the decline in bacterial numbers and NET values did not return to baseline for approximately 50 more hrs. Moreover, this resolution correlated with oscillating Salmonella numbers during the waning 25 to 60 hrs of the infection.

Others have observed oscillatory patterns in host responses to other types of bacteria in mouse infections [ ]. Therefore, we were reassured that the simulation captured the essence of a real infection.

As shown in Fig 6 , the CRP level rose initially after the infection and waxed and waned for another 2 to 3 days. On the 4 th day after infection, CRP levels diminished sharply and damaged hepatocytes began their recovery, similar to the CRP pattern reported in a clinical study [ ].

Counts of different variables agents were measured at each simulation time point of one selected simulation. Detectable hepatocyte damage began at simulation step 10 10 hrs post infection , and a significant increase in hepatocyte damage was observed beginning at stimulation step 15 15 hrs post infection. Hepatocyte damage was persistently observed for 7 days. As the persistent infection proceeded, a large area of hepatocyte damage, which would translate to liver damage in an animal model, was observed Fig 7.

These data are consistent with the idea that a persistent infection will induce a higher mortality rate compared to a healing response because acute tissue damage is more detrimental to the host Fig 6B. Remarkably, we observed that oscillations in agent counts were damped when calculated mean values of the agent counts for simulation replications Fig 6.

However, a significant elevation in phagocytic cells Fig 8E, 8H and 8I and inflammatory cytokines was observed Fig 8J, 8K and 8L compared to the healing and the persistent infection responses, causing severe hepatocyte damage that could lead to death Fig 8B. This made it difficult to accurately predict outcomes in this type of HIR. These data suggest that a hyperinflammatory response could lead to a higher mortality rate compared to a persistent infection because of the acute and severely damaged hepatocytes observed.

The last pattern of HIR that we observed was characterized by progressively increasing Salmonella counts.

Under this condition, Salmonella and inflammatory cytokines continued to rise as the simulation proceeded. Specifically, the simulation stopped under the condition that no more healthy hepatocytes existed. We only calibrated the data of organ dysfunction for the first 24 hrs of HIR because healthy hepatocytes died out at 24 hrs of simulation in some replications.

The acute rise and a slow decrease in CRP levels observed in our model were consistent with CRP concentration patterns identified in patients with septic shock [ ]. To conclude, we found that a healing response, where Salmonella , other phagocytic cells, and inflammatory cytokines quickly fell below threshold levels, was more likely to occur when the initial Salmonella load was low.

We identified a persistent infection pattern if inflammatory responses were active characterized as when Salmonella and inflammatory cell levels oscillate during infection.

However, if the initial Salmonella load was high, a hyperinflammatory response or organ dysfunction was most likely to occur, leading to the death of infected individuals. In addition, when these simulated results were compared to experimental data, the simulations paralleled indicator patterns reported in actual mouse experiments [ , , , — , ].

It also became clear that predicting a final outcome from the emerging dynamic patterns of HIR became more difficult when initial Salmonella loads were above counts See Section 3.

To examine how the model behaved when parts of the immune response were absent we ran IMMABM simulations with and without acquired immunity using initial Salmonella doses ranging from counts to counts.

In these experiments, doses were increased in increments of counts with replications per dose for a total of replications in the IMMABM. In the absence of acquired immune components, HIR outcomes clearly skewed toward a healing response at doses less than counts. However, as the initial Salmonella doses increased, it became clear that the dynamic patterns of HIR could diverge in the health outcomes healing response vs.

For example, when the initial Salmonella load was counts, all four dynamic patterns of HIR could emerge. However, when the initial Salmonella count exceeded counts, only hypothetical death status hyperinflammatory response or organ dysfunction was identified from IMMABM simulations.

In order to compare potential survival and mortality rates of HIR under various initial Salmonella challenge loads, we generated a probability histogram that ended with the healing response, persistent infection, hyperinflammatory response, or organ dysfunction of HIR against various Salmonella initial loads Fig The probability of HIR ending in each possible outcome clearly changes as the dose increases from counts to counts Fig These data show the importance of innate immunity in the control of Salmonella infections [ , ].

However, the likelihood of organ dysfunction dropped compared to when acquired immune components were missing Fig It took substantially higher challenge doses to induce more severe infections S2 Fig. Therefore, the model seemed to accurately reflect the relative contributions of innate and acquired immunity during Salmonella infections [ 39 , , ].

As Salmonella initial loads increased from to , the chance of a hyperinflammatory response was significantly higher than organ dysfunction S2 Fig. Although both conditions can be detrimental to the host, the model reflects subtle changes in the infection and the rapid and overwhelming pro-inflammatory response induced by a high initial loads of Salmonella.

Interestingly, the overwhelming pro-inflammatory response damages hepatocytes at an early stage of HIR results in few Salmonella replications within hepatocytes. It is also interesting to note that in other Gram-negative bacterial infections, the absence of T cells results in prolonged neutrophila hyperinflammtion in the lung [ 2 ].

Again, suggesting that the model is beginning to reflect the in vivo situation. Experimental data show that cellular and soluble mediator interactions and concentrations change and their levels are dependent on location and time. For example, the Salmonella killing rate by one neutrophil can range from 2. Consistent with its embedded stochastic nature, the IMMABM allowed us to determine the probability of each possible outcome in individuals, thereby allowing reasonable predictions of HIR outcomes.

In contrast, lower initial challenge doses were more likely to be identified as healing response or persistent infection.

As described in Section 3. Similar to our simulated results, persistent elevation of HMGB-1 and CRP was also observed in experimental studies [ , , ]. Although it is clear that MDMI polarization is common in bacterial infections [ ], it is less clear if macrophage polarization is associated with host dysfunctional responses.

Therefore, it is possible that our simulated data reflect the in vivo ambiguity. Refinement of the model will be necessary to help resolve this. We designed an experiment using a hypothetical antimicrobial agent i. Simulated data showed that the treatment effectiveness was highly correlated with treatment start time during the simulation Fig Specifically, antimicrobial agents caused significant improvement in the survival rates when started during the first hour after infection beginning of HIR.

Interestingly, current recommendations are to administer appropriate antibiotics within 1 hour of a diagnosis of severe sepsis or septic shock [ , ]. Our simulated results demonstrated that effectiveness of anti-agent treatments has a specific time window. This could lead to organ dysfunction because there would be insufficient numbers of phagocytes to ingest and kill intruding Salmonella. This indicates that higher amounts of anti-agents could be necessary during treatment as Salmonella infection levels increase.

The insights provided by the therapy-directed experiment suggest that various doses of anti-agent treatment e. Most importantly, the IMMABM was validated through a series of comparisons between simulated results and experimental studies.

Four distinct dynamic patterns healing response, persistent infection, hyperinflammatory response, and organ dysfunction were identified during the IMMABM simulation. One significant finding from the simulations was that the outcomes of a HIR were highly correlated to the initial Salmonella counts. As the initial counts of Salmonella increased, HIR had a higher probability to end with hyperinflammatory or organ dysfunction responses. In the therapy-directed experiment, we observed that antimicrobial intervention significantly improved the survival rates during the first hour.

T cell activation occurs within 24 hrs of HIR in vivo [ ]. We found that incorporation of T cells and other acquired immune components could induce changes to the course of the infection. Indeed, in immunocompetent mice, the hyperinflammatory response that we identified in the simulations paralleled observations made during mouse sepsis [ ].

In addition, we found that antibodies released during HIR failed to significantly affect organ dysfunction based on the release rate of antibodies and the binding amount of antibody to one Salmonella we calibrated [ — ].

During dose response simulations, we observed HIR is correlated to initial loads of Salmonella. Recent studies found that a high bacterial load is significantly associated with worse outcomes [ ]. Our study reported the first probability histogram that ended with various patterns under a range of initial Salmonella loads. The probability histograms describe the relationship between probability of ending with each HIR pattern and Salmonella challenge levels. Step 3: Click on the Update Now button next to the flagged device driver, i.

Performing the aforementioned steps not only takes care of all USB 3. USB 3. You can manually download and install the USB 3. Note: To download USB 3. Before downloading drivers manually, make sure that the driver version is compatible with the version of Microsoft Windows and processor type. Below are the step-by-step instructions on how to install the latest Intel USB 3. Step 2: Extract the downloaded zip file i. Step 3: Now, right-click on the Windows logo icon and select Device Manager.

The first USB 3. See image below. Step 6: Now, right-click on your USB 3. Step Choose the. Step Lastly, restart your computer to finish the driver installation process. This manual method requires a lot of time, patience, and technical knowledge.

In addition to this, downloading and installing the wrong drivers can make problems even worse. Here are the steps that you need to follow for updating the driver for USB 3. So, modifying the power settings can resolve the USB 3. Here are the steps that you need to follow to modify the Windows 10 USB power settings:.

Step 2: Then, type devmgmt. Alternatively, you can also make a click on the OK option. Step 3: After this, when the Device Manager window appears, scroll down to choose Universal Serial Bus Controllers and double click on it to expand the list.

Step 6: Tick-mark the checkboxes presented before the Allow the computer to turn off this device to save power.