Latest AI and User-Annotated Protocols for Enhanced Reproducibility

All in vitro α-glucosidase inhibition assays were performed in a 96-well microplate format. The α-glucosidase inhibition assay reported elsewhere was customized to suit the experimental needs of the current work [16 (link)]. The assay mixture contained 50 mM sodium phosphate buffer (pH 6.8), 20 mU of α-glucosidase (EC 3.2.1.20) from Saccharomyces cerevisiae (Sigma Chemical Co., St. Louis, USA), and 1.0 mM of p-nitrophenyl-α-D-glucopyranoside (PNP-G) substrate. All target compounds were dissolved in DMSO with the assay mixture containing DMSO at a final concentration of 3%. All compounds were first incubated with the enzyme at 37 °C for 15 min followed by commencement of the enzyme reaction by adding the substrate. The reaction was continuously monitored at 400 nm in SpectraMax Plus 384^® microplate reader (Molecular Devices, CA, USA). The assay also included separate, parallel sets of control experiments containing no compounds as negative control, and acarbose as positive control. All assays were performed as four individual experiments out of which the values of mean and standard error of mean (SEM) were calculated.

In seismology and earthquake engineering, resilience is an approach that introduces the time dimension to explain the post-seismic event recovery phase which covers the scopes beyond the single building to communities including both essential and non-essential systems [1] . Generally, seismic resilience is defined as the ability of a structure or a system to continue operating as normal or probably extends its serviceability after the initial damage has been repaired [2] . The concept of seismic resilience plays a critical role in the phases of rescue and recovery after seismic events by assessing the post-event functionality of the affected systems or structures. As described by Ahern [3] , the concept of resilience changes the structural design requirements from ‘fail-safe’ to ‘safe-to-fail’, for further explanation, a system or a structure will not fail at a prescribed disturbance level or the failure of the system or structure at a higher disturbance level will be acceptable and allowable for a recovery phase at an optimal combination of repair time and repair cost.
In the early decades, community resilience was developed and evaluated qualitatively and conceptually, which was less substantial and efficient. It is possible to use the community resilience index as a starting point for tracking changes in several areas of the community through time, including socio-demographics, economy, the environment, organizational structures, and culture. Due to limited political and planning actions, the idea of community resiliency is rarely used as a tool in earthquake risk management and mitigation. This results in a defective, ineffective reconstruction procedure because of a lack of consistent, and adequate mitigating solutions. There have been various classic frameworks established recently to analyze the resiliency of communities, effective and beneficial for both single systems and the entire community. For instance, You et al. [4] , established a framework that has to link between the seismic performance of structures, and the resiliency of communities. Figure depicts the framework Fig. 1.
Indeed, over the last few years, a great number of quantitative frameworks have been developed in order to evaluate the structural resilience against seismic tremors for structures that already exist in a variety of countries. Cimellaro et al. [5] offered multiple frameworks to characterize resilience quantitatively relying on an analytical approach that can be used for technological and organizational considerations. Cimellaro et al. introduced a comprehensive model to quantify the seismic resiliency of a hospital system that incorporates both loss estimation and recovery models that can be employed for critical facilities. For essential facilities, Cimellaro et al. [6] developed a comprehensive model for quantifying seismic resiliency that includes both loss estimation and restoration models. Hassan et al. [7] developed a framework for a hospital with 6- storeys that would be used to evaluate the post-earthquake functionality of the hospital system, by comprising both quantity (space availability, personnel availability, and supplies availability) and quality (satisfaction with medical treatments) portions. Nevertheless, a methodology is developed by Shang et al. [8] that contains four phases which is used to estimate the seismic resilience of hospital systems, including a seismic risk analysis, seismic hazard analysis, fragility analysis, and the determination of seismic resilience. In addition, Yarveisy et al. [9] developed straightforward resilience evaluation criteria for the purpose of quantifying and evaluating resilience based on the concepts of reliability and serviceability.
It is worth noting that, a building's resilience can be defined as its ability to withstand external threats while also regaining its functionality following damage or destruction. Structural resilience has been explored by Bruneau et al. [10] by identifying four major traits. The 4R attributes stand for resiliency, rapidity, redundancy, and resourcefulness. As stated by Lu et al. [11] , there are two main known approaches to evaluate a building's structural resilience: risk and resistance analysis of structures; and rating of structural vulnerability. Risk and resistance analysis of buildings is the first method that has been developed. It is common practice to base a structure's design and information on hazards such risk and design loadings on the environment, structural element types, building materials, and GIS data for the structure. Consequently, the building's structural resilience can be estimated using the design in 1 formation. The United States Green Building Council (USGBC) recommends the Leadership in Energy and Environmental Design (LEED) analysis and planning for resilience. This process necessitates a hazard assessment of the construction project. The Building Resilience Rating Tool (BRRT) was also developed by the Insurance Council of Australia (ICA), and it was based on this methodology [12] . Moreover, BRRT rates building resilience by recognizing probable dangers. As a result, the building's materials and structural types are analyzed to determine its level of risk. On the other hand, the grading of structural resilience under a specified hazard scenario is the second method now in use. The structure's resistance can be graded or given stars based on indicators related to the 4R and other attributes that are discovered when a threat is detected. Grading systems for buildings have been proposed by the Resilience-Based Earthquake Design Initiative (REDi) as well as by the US Resiliency Council (USRC). These rating systems are based on the buildings' ability to withstand earthquakes.
The evaluation of the seismic resilience index of a structure is an essential initial step in the process of formulating a plan for repairs or upgrades. Using the functionality curves, it is possible to effectively suggest or implement a repair plan or retrofitting approach based on the results of the resilience assessment and the amount of time it takes for the system to recover. For instance, Samadian et al. [13] carried out research in Iran with the purpose of determining the seismic resilience of both newly constructed and previously used RC school buildings. The research was examined using vulnerability and fragility curves. The fundamental purpose of this research is to employ a new method that concentrates on the economic situations of the provinces through the utilization of vulnerability analysis in order to assess the losses that were caused by earthquakes. According to the findings, the SRI had a functional decline as the magnitude of the threat levels was raised higher. Titi and Biondini [14] conducted an investigation into the dependability of concrete frame structural systems that were negatively damaged by corrosion. The findings demonstrated that the constructed structures that were intended to have the same degree of functioning could display differing levels of seismic resistance over a predetermined amount of time as a direct result of the environmental variables in their immediate surroundings. Other investigations were carried out with the purpose of determining the levels of resilience and SRI possessed by various types of buildings, for example, Banerjee and Chandrasekaran [15] explored the seismic resilience of bridges when they were subjected to the influence of various hazards, whereas Alipour and Shafei [16] assessed the seismic resilience of transportation networks.
In this paper, a research methodology is applied for evaluating the seismic resiliency of damaged reinforced concrete (RC) buildings in Malaysia under Ranau ground motion at seismic intensity measure (VIII) in which a soft story failure mechanism phenomenon is appeared. An RC school building with four floors and an RC hospital building with five stories were chosen for this research so that the suggested seismic resilience index (SRI) methodology could be applied. Nevertheless, neither of these models were built to withstand the effects of seismic loading. Moreover, there are three different clusters of existing RC structures in the study area: low-, mid- and high-rise. These three building types were chosen for the case studies in order to assess the SRI and present its findings graphically using the Geographical Information System (GIS).

The total antioxidant capacity of the fractions was determined by phosphomolybdate method using ascorbic acid as a standard [21 ]. An aliquot of 0.1 ml of sample solution was mixed with 1 ml of reagent solution (0.6 M sulphuric acid, 28 mM sodium phosphate and 4 mM ammonium molybdate). The tubes were capped and incubated in a water bath at 95°C for 90 min. After the samples had cooled to room temperature, the absorbance of the mixture was measured at 765 nm against a blank. A typical blank contained 1 ml of the reagent solution and the appropriate volume of the solvent and incubated under the same conditions. Ascorbic acid was used as standard. The antioxidant capacity was estimated using following formula:
$\text{Antioxidant}\text{effect}\left(\%\right)=\left[\left(\text{control}\text{absorbance}-\text{sample}\text{absorbance}\right)/\left(\text{control}\text{absorbance}\right)\right]\times 100$

Pellets containing extracellular vesicles isolated from Ishikawa cells (i.e., exosomes) were vortexed and resuspended in 1 mL of distilled water. The solution was further purified using dialysis tubing cellulose membrane (molecular weight cut-off = 14,000) against 1 L × 3 distilled water for 24 h. A 20 μL solution was then placed on the center of the scanning electron microscope (SEM) sample holder (ϕ = 15 mm) and dried in a ventilation hood for 30 min. A 20 μL volume of ethanol was then directly placed on the dried sample to dehydrate the exosomes. After evaporating the ethanol at room temperature for 30 min, the sample holder was placed in a vacuum chamber for 2 min to eliminate any outgassing from the exosomes and water. To generate a conductive surface, a 10-nm gold coating was applied by sputtering (ion sputter coater) prior to imaging by SEM (SNE-4500 M, South Korea). SEM was used at 10.0 kV and the images obtained were assessed using the Image J software (http://imagej.nih.gov/ij/) software for analysis of exosome size.

CT qPCR was conducted by the use of primers targeting the single-copy ompA gene, coding for the major outer membrane protein (MOMP), as described by Jalal et. al. [17 (link)]. QPCR was performed on a 7900HT Real-Time PCR System (Applied Biosystems, Foster City, California), in a total volume of 25 μl per reaction, consisting of 10 μl purified DNA and 15 μl reaction mixture. The qPCR reaction mixture contained 12.5 μl Absolute qPCR Rox Mastermix (Thermo Scientific, Waltham, USA) and 2.5 μl primer/probe mix consisting of 840 nM forward and reverse primer and 100 nM probe. The amplification conditions were 15 minutes of initial activation at 95°C, followed by 42 cycles of 95°C for 15 seconds and 60°C for 60 seconds. Cycle threshold-values were entered into a master calibration curve to determine chlamydial load in log CT/ml, as described elsewhere [18 (link)].

Cell proliferation was measured using CCK-8 according to a previous study [21 (link)]. In brief, HaCaT cells (2 × 10⁴) were added into 96-well plates. After grown for 0, 24, 48, 72 h, 10 μL CCK-8 (Beyotime) was infused, and cells were cultured for 2 h. The optical density value at 450 nm was detected with Epoch2 microplate reader (Biotek, Winooski, VT, USA).

HaCaT cells were seeded onto different samples with a density of 2 × 10⁴/well. After 1, 3, and 5-day of seeding, cell proliferation was evaluated using Cell Counting kit-8 (Dojindo, Japan). At the setting time point, culture medium was replaced with 500 μl of fresh medium containing 50 μl of CCK-8 reagent, and the samples were incubated at 37°C for 2 h. The optical densities (OD) of the mixed solution were measured using a SpectraMax microplate reader (Molecular Devices, United States) with wavelengths of 450 nm. Each group contained 3 samples, and the mean value served as the final result.

An LC-MS method previously described [21 (link),22 ] was used for the identification of individual polyphenols in the HI extracts. The 18 external standards were: apigenin, caffeic acid, caftaric acid, chlorogenic acid, ferulic acid, fisetin, gentisic acid, hyperoside, isoquercitrin, kaempferol, luteolin, myricetin, patuletin, p-coumaric acid, quercetin, quercitrin, rutoside, and sinapic acid (Table S1). In brief, the chromatographic separation was performed on a reverse-phase analytical column (Zorbax SB-C18, 100 mm × 3.0 mm i.d., 3.5 µm particles) with a mixture of methanol/acetic acid 0.1% (v/v) as mobile phase and a binary gradient. The elution started with a linear gradient, beginning with 5% methanol and ending at 42% methanol at 35 min, isocratic elution followed with 42% methanol for the next 3 min, rebalancing with 5% methanol in the next 7 min. The flow rate was 1 mL/min, the injection volume was 5 µL, and the column temperature was 48 °C. The detection procedure was performed on both UV and MS mode. The UV detector was set at 330 nm until 17 min (for the detection of polyphenolic acids), then at 370 nm until the end of analysis time (for the detection of flavonoids and their aglycones). The MS system operated using an electrospray ion (ESI) source in negative mode (capillary 3000 V, nebulizer 60 psi (nitrogen), dry gas temperature 360 °C, and dry nitrogen gas at 12 L/min).
Another LC-MS method (LC-MS method II) previously described [23 (link)] was used to detect the other six polyphenols (epicatechin, catechin, syringic acid, gallic acid, protocatechuic acid, and vanillic acid) in HI extracts (Table S2). The chromatographic separation was accomplished on the same analytical column and in the same chromatographic conditions as mentioned before but with a slightly different binary gradient (start: 3% methanol; at 3 min: 8% methanol; at 8.5 min: 20% methanol; at 10 min: rebalance column with 3% methanol). The detection of the compounds was performed on MS mode. All identified polyphenols were measured both in the HI non-hydrolyzed and hydrolyzed extracts (equal amounts of extract and 4 M HCl kept 30 min on 100 °C water bath) on the basis of their peak areas and comparison with a calibration curve of their corresponding standards. The results were expressed as micrograms of phenolic compounds per dw of involucre (μg/g dw).

30 university students were recruited through convenience sampling for the study. All the participants were healthy and free of diabetes mellitus, orthopedics, and neuromuscular diseases, and with no obvious structural asymmetry of bilateral lower limbs. The study was conducted in the Motion Analysis and Motor Performance Laboratory at the University of Virginia (UVA). The Human Investigation Committee of UVA monitored and approved all procedures of the present study. Consent was obtained for each participant enrolled (HSR#:16853).
The study protocol consisted of 3-dimension (3D) gait analysis for 3 walking conditions: barefoot, in MBFT shoes, and in neutral shoes (Figure 1). The OESH® (La Vida, from the OESH Barefoot Technology®) shoes that have completely flat soles, with no arch support and no heel lift, which aim to mimic barefoot, were used as the MBFT shoe condition in this study. A current widely used neutral running shoe (Brooks®, Radius 06) was used as the neutral shoe condition. The order of walking conditions tested was randomly decided by coin flipping of each enrolled individual.
Enrolled subjects were instructed to walk along a 10-meter laboratory walkway at their self-selected comfortable walking speed (CWS), wearing a Plug-in-Gait full body 37-marker set (Vicon, Oxford, UK). 3D kinematic and kinematic data were collected with an 8-camera Vicon Motion Analysis System (Vicon, Oxford, UK) at 120 Hz, and the data of contact reaction forces was collected using 4 in-ground force plates (Kistler, Switzerland and Betec, OH) at 1080 Hz. The acceleration and deceleration at the force plate were well controlled in each trial, and trials with walking speeds close to barefoot walking tests were chosen from the OESH shoe and neutral shoe walking tests; at least 5 successful trials which met these criteria were recorded for each subject.
Gait kinetics and kinematics variables were computed using Vicon's full body Plug-in-Gait models [12 (link)]. All variables were normalized to the stance phase. To evaluate gait symmetry, the equations for symmetry index (SI) below developed by Nigg et al. [11 (link)] were applied: $\begin{array}{c}\mathrm{S}\mathrm{I}=\underset{t=t\mathrm{1}}{\overset{t\mathrm{2}}{{\displaystyle \int }}}A\left|Xr\left(t\right)-Xl\left(t\right)\right|dt,\\ \end{array}$ $\begin{array}{c}A=\frac{2}{\mathrm{r}\mathrm{a}\mathrm{n}\mathrm{g}\mathrm{e}\left(Xr\left(t\right)\right)+\mathrm{r}\mathrm{a}\mathrm{n}\mathrm{g}\mathrm{e}\left(Xl\left(t\right)\right)},\\ \end{array}$ where Xr(t) and Xl(t) are specific variables recorded for the right leg or the left leg at the time t and t1 and t2 refer to the times at heel-contact and toe-off, respectively. Therefore, evaluating gait symmetry with this SI consecutively incorporates the data of entire stance phase rather than discrete time points. In (1), A is used to normalize the data over range, and range is used instead of the mean of the data so that nonsimilar gait parameters could be compared. For any given variable, SI = 0 means perfect symmetry, while on the contrary, the larger the SI value is, the less symmetric the gait it indicates.
As per previous research, the variables of joint angle, joint force, and joint moment [5 (link), 13 (link)] were selected for the calculation of SI, using the software of MATLAB® (The MathWorks, Inc., Natick, MA). According to Nigg's methodology [11 (link)], SI was calculated separately for each of the three joints of the lower limb (hip/knee/ankle) and for each of the three motion planes (sagittal/transverse/frontal) and was also calculated jointly with data from all the 3 lower limb joints and all the 3 motion planes, so that gait symmetry can be evaluated from both categorical and global perspectives.
All statistics analysis was performed with SPSS 20.0 for windows (SPSS, Inc., Chicago, IL). Continuous variables with normal distribution were presented as mean ± standard deviation. Leven's test was conducted to test the heterogeneity of data and log-transformed the data if necessary. One-way ANOVA was used for the intergroup comparisons; nonparametric tests were applied if heterogeneity of the data was not fulfilled by log transformation. p < 0.05 was considered as statistically significant.

Peripheral blood mononuclear cells (PMNCs) were isolated from the blood samples of the individuals who consented to their participation in the study. They were then reprogrammed to iPSCs by using a transient expression method (nucleofection) involving three plasmid vectors (MOS, MMK and GBX) under feeder-free/xeno-free culture on 4D Nucleofector (Lonza, Basel, Switzerland) [12 (link),22 (link),23 (link),24 (link)]. Generated iPSCs were characterized by immunocytochemistry for pluripotency markers, including Nanog, OCT4 and TRA-1-60, flow cytometry, and karyotyping. Established cell lines were cultured (250 K cells/well in 6 well plate) on vitronectin-coated tissue culture plates in Essential-8 (E8) medium with 10 uM Rock inhibitor (Y-27632) during seeding and then maintained in E8 medium till 80–90% confluency [12 (link),23 (link)].

Adipose tissue pads were digested in 1.5 mg/ml collagenase type I (Invitrogen) in DMEM for 40 min at 37 °C in CO₂ incubator without shaking. The digested mixture was filtered through a 100μm cell strainer (BD Biosciences) and centrifuged at 800 xg for 10 min. The upper layer (adipocytes) and the pelleted cells (SVF) were collected for further analysis.

Epididymal adipose tissues (eAT) from ob/+ mice and ob/ob mice or chow and high-fat diet fed mice were subjected to collagenase digestion to obtain adipocytes and the stromal vascular fraction (SVF) as described [30 (link)]. Briefly, finely cut eAT pieces were incubated for 45 min at 37 °C with collagenase (1 mg/mL) in Krebs-Ringer medium, buffered with bicarbonate and HEPES (KRBH: 120 mM NaCl, 4.8 mM KCl, 2.5 mM CaCl₂, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄, 15 mM NaHCO₃, 30 mM HEPES, pH7.4), and containing 0.3 mM glucose and 1% BSA. After digestion, eAT cell suspension was filtered through nylon screen (pore size 250 mesh) in 50 mL polypropylene conical tube. The medium under the floating adipocytes was collected and centrifuged (1000× g 10 min) to obtain a cell pellet corresponding to the SVF. The floating adipocytes were washed three times by flotation with KRBH without BSA. The cell pellet corresponding to the SVF was washed by resuspension in KRBH without BSA followed by centrifugation. The adipocytes and SVF were stored at −80 °C for further analysis.

iWAT from 8–12-week-old male mice (diet: regular chow) was isolated and digested at 37 °C for 30–45 min in Krebs-Ringer-Hepes-bicarbonate buffer (KRH buffer, 1.2 M NaCl, 40 mM KH₂PO₄, 20 mM MgSO₄, 10 mM CaCl₂, 100 mM NaHCO₃, and 300 mM HEPES) containing 3.3 mg/ml collagenase I (Sigma-Aldrich). Once the tissue was fully digested, 10 ml of KRH buffer was added to prevent further collagenase activity. Cells were then filtered through a 250 μm cell strainer. Ten min later, the floating top layer containing mature adipocytes was collected. Mature adipocytes were washed twice with KRH buffer containing 5 mM EDTA. After collection of the mature adipocytes, the digested tissue suspension was filtered through a 100 μm cell strainer, followed by centrifugation at 400 g for 5 min at room temperature. The cell pellet was re-suspended in Dulbecco’s PBS (dPBS) and filtered through a 40 μm cell strainer. The cell suspension was centrifuged again as described above to obtain a pellet containing the stromal vascular fraction (SVF) cells.

Murine HL-1 cardiomyocytes (HL-1), immortalized using the simian virus SV40 T-antigen under the control of an atrial natriuretic factor (ANF) promotor [46 (link)], were generously provided by William C. Claycomb (Louisiana State University, New Orleans, LA, USA) and used at passages 18–44. Cells were cultured in Claycomb medium with 10% HL-1 Cell Screened FBS (Merck), 100 U/mL penicillin, 100 µg/mL streptomycin, 2 mM L-Glutamine (Gibco, Waltham, MA, USA) and 100 µM norepinephrine, and cultured on 0.02% gelatin/5 µg/mL fibronectin-coated (Sigma-Aldrich) cell culture flasks at 37 °C and 5% CO₂.

Synthetic DMPC (1,2-dimyristoyl-sn-glycero-3-phosphocholine), D54-DMPC (1,2-dimyristoyl-d54-sn-glycero-3-phosphocholine), DMPG (1,2-dimyristoyl-sn-glycero-3-phospho-(1′-rac-glycerol)), D54-DMPG (1,2-dimyristoyl-d54-sn-glycero-3-[phospho-rac-(1-glycerol)] (sodium salt)) and DMPE-PEG (1,2-dimyristoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (ammonium salt)) were purchased from Avanti Polar Lipids. Lipid films where prepared by dissolving the lipids in a methanol: chloroform solution to a 1 : 3 volume ratio, followed by solvent removal under a stream of nitrogen flow. The vials where then left under vacuum for at least one hour to ensure complete removal of organic solvents. Lipid films were then kept at −20 °C until use.

DMPC, DMPG, and DOPC were purchased from Larodan Fine Chemicals AB (Malmö, Sweden). DOPS and the deuterated d₅₄-DMPC and d₅₄-DMPG were purchased from Avanti Polar Lipids (Alabaster, Alabama, USA).
Lipid stocks were prepared by dissolving dry lipids in chloroform or chloroform:methanol (9:1 v/v) at 10–30 mM. Mixtures were prepared from stocks at the desired molar ratios, followed by solvent evaporation under a stream of nitrogen and lyophilizing overnight at -52°C. The dried lipids were stored at -20°C or used directly for liposome preparation.
Liposomes were prepared by mixing dried lipids with water or HBS at 10–15 mM, followed by inverting at ambient temperature for at least 3 h. Multilamellar vesicles (MLV) were prepared by freeze-thaw cycles in liquid N₂ and a warm water bath and vortexing. The cycle was performed 7 times in total. Large unilamellar vesicles (LUV) were prepared by passing fresh MLVs through a 0.1-μm membrane 11 times at 40°C. SUVs were prepared by ultrasonication of fresh MLVs using a probe tip sonicator (Sonics & Materials Inc. Vibra-Cell VC-130) until clarified. All lipid preparations were immediately used in experiments.

The study followed an open-label, randomized design. Patients with chronic pain from fibromyalgia were randomized in a 1:1:1 ratio to self-titrate 5 mg oral oxycodone sustained-release tablets (Mundipharma Pharmaceuticals BV, The Netherlands), 150 mg of inhaled cannabis (Bediol, Bedrocan International BV, The Netherlands) containing 6.3% THC (9.5 mg) and 8% CBD (12 mg) with a specific terpene profile (including myrcene, α₂-pinene, terpinolene, β-caryophyllene, cis-ocimene, α-terpineol, all >0.5 mg/g; see https://bedrocan.com/wp-content/uploads/bediol-terpene-profile.pdf) or their combination for 6 weeks with a 6-week follow-up. Patients who were treated with cannabis received capsules that contained 150 mg of cannabis of the Bediol variety. These capsules could be placed in a hand-held vaporizer, the Mighty Medic (Storz & Bickel GmbH & Co., Germany), which was provided to patients for home use. The vaporizer heats the plant material to allow the conversion of THC and CBD acids into active THC and CBD for inhalation. The patients inhaled cooled cannabis vapor from the vaporizer in sessions lasting 3–5 min. The study was performed at a single center (LUMC) where patients received their medication; however, all patients were treated at home. The cannabis formulation and dosage used in this study were based on our earlier study comparing the effects of three cannabis varieties, which found that the Bediol variety was most effective in reducing fibromyalgia pain (4 (link)).
Randomization was performed by the study team using the Castor data capture system (www.castoredc.com) on the day prior to dosing. Patients assigned to oxycodone treatment could use up to two (5 mg) oxycodone tablets per day during week 1 of treatment, increasing to up to four oxycodone tablets per day during weeks 2–6. Patients assigned to cannabis treatment were instructed by the study team to use up to three (150 mg) cannabis capsules per day in week 1, increasing to up to five capsules in weeks 2–6. Patients assigned to the combination of cannabis and oxycodone were allowed to use up to two oxycodone tablets and up to three cannabis capsules per day in week 1, increasing to up to four oxycodone tablets and up to five cannabis capsules per day during treatment weeks 2–6. Medication was prepared by the pharmacy and provided to patients through the study team.