Predicting combinative drug pairs via multiple classifier system with positive samples only

BACKGROUND AND OBJECTIVE

Due to the synergistic effects of drugs, drug combination is one of the effective approaches for treating complex diseases. However, the identification of drug combinations by dose-response methods is still costly. It is promising to develop supervised learning-based approaches to predict potential drug combinations on a large scale. Nevertheless, these approaches have the inadequate utilization of heterogeneous features, which causes the loss of information useful to classification. Moreover, they have an intrinsic bias, because they assume unknown drug pairs as non-combinations, of which some could be real drug combinations in practice.

METHODS

To address above issues, this work first designs a two-layer multiple classifier system (TLMCS) to effectively integrate heterogeneous features involving anatomical therapeutic chemical codes of drugs, drug-drug interactions, drug-target interactions, gene ontology of drug targets, and side effects. To avoid the bias caused by labelling unknown samples as negative, it then utilizes the one-class support vector machines, (which requires no negative instance and only labels approved drug combinations as positive instances), as the member classifiers in TLMCS. Last, both a 10-fold cross validation (10-CV) and a novel prediction are performed to validate the performance of TLMCS.

RESULTS

The comparison with three state-of-the-art approaches under 10-CV exhibits the superiority of TLMCS, which achieves the area under the receiver operating characteristic curve = 0.824 and the area under the precision-recall curve = 0.372. Moreover, the experiment under the novel prediction demonstrates its ability, where 9 out of the top-20 predicted combinative drug pairs are validated by checking the published literature. Furthermore, for each of the newly-validated drug combinations, this work analyses the combining mode of the member drugs and investigates their relationship in terms of drug targeting pathways.

CONCLUSIONS

The proposed TLMCS provides an effective framework to integrate those heterogeneous features and is trained by only positive samples such that the bias of taking unknown drug pairs as negative samples can be avoided. Furthermore, its results in the novel prediction reveal five types of drug combinations and three types of drug relationships in terms of pathways.

Vial washers for the compounding pharmacy: ensuring preparation safety

The safety and effectiveness of customized formulations have always been priorities for compounding pharmacists, perhaps even more so since a tragic outbreak of meningitis in 2012 was traced to preparations from the New England Compounding Center in Framingham, Massachusetts. Since that time, pharmaceutical compounding in the U.S. has been the focus of renewed interest from the public and from government organizations and regulatory bodies created to ensure patient safety. As a result, responsible compounders have responded to ensure--to even greater levels--the purity and safety of the formulations they prepare. One tool useful in doing so is the mechanical vial washer, which can produce cleaner vials (and/or ampules, syringes, and cartridges, depending on the washer model) more consistently than does washing vessels by hand. In this article, several such washers appropriate for use in a compounding pharmacy are profiled, and a pharmacist's experience in using one of those models is described.

Pharmaceutical dust exposure at pharmacies using automatic dispensing machines: a preliminary study

Automatic dispensing machines (ADMs) used in pharmacies concentrate and dispense large volumes of pharmaceuticals, including uncoated tablets that can shed dust. We evaluated 43 employees' exposures to pharmaceutical dust at three pharmacies where ADMs were used. We used an optical particle counter to identify tasks that generated pharmaceutical dust. We collected 72 inhalable dust air samples in or near the employees' breathing zones. In addition to gravimetric analysis, our contract laboratory used internal methods involving liquid chromatography to analyze these samples for active pharmaceutical ingredients (APIs) and/or lactose, an inactive filler in tablets. We had to choose samples for these additional analyses because many methods used different extraction solvents. We selected 57 samples for analysis of lactose. We used real-time particle monitoring results, observations, and information from employees on the dustiness of pharmaceuticals to select 28 samples (including 13 samples that were analyzed for lactose) for analysis of specific APIs. Pharmaceutical dust was generated during a variety of tasks like emptying and refilling of ADM canisters. Using compressed air to clean canisters and manual count machines produced the overall highest peak number concentrations (19,000-580,000 particles/L) of smallest particles (count median aerodynamic diameter ≤ 2 μm). Employees who refilled, cleaned, or repaired ADM canisters, or hand filled prescriptions were exposed to higher median air concentrations of lactose (5.0-12 μg/m(3)) than employees who did other jobs (0.04-1.3 μg/m(3)), such as administrative/office work, labeling/packaging, and verifying prescriptions. We detected 10 APIs in air, including lisinopril, a drug prescribed for high blood pressure, levothyroxine, a drug prescribed for hypothyroidism, and methotrexate, a hazardous drug prescribed for cancer and other disorders. Three air concentrations of lisinopril (1.8-2.7 μg/m(3)) exceeded the lower bound of the manufacturer's hazard control band (1-10 μg/m(3)). All other API air concentrations were below applicable occupational exposure limits. Our findings indicate that some pharmacy employees are exposed to multiple APIs and that measures are needed to control those exposures.

From Tonic-cups to Bitter-cups: Kwasi bita beker from Suriname Determination, past and present use of an ancient galenic artefact

In the main markets of Paramaribo (Suriname), many stands offer what is locally called "Bitter-cups", or "Kwasi bita beker", small footed-cups, roughly carved from a whitish wood. The use of these cups is strictly medicinal and it seems to be restricted to Suriname, as they are not found in neighbouring countries (Guyana, French Guiana). The aim of this study was to identify the botanical origin of Bitter-cups still in use in the Saramaka traditional medicine (as information from field people was controversial), and document the ethnopharmacology of this original galenical artefact. Microscopic and high performance liquid chromatography (HPLC) analyses were carried out on Bitter-cup, and anatomical criteria (marginal parenchyma band, size of intervessel and vessel-ray pits, rays width and rays composition, vessels clustering, frequency and size of parenchyma pits) together with HPLC profiles of the macerates showed that the wood cup was similar to Quassia amara L. (Simaroubaceae) wood. Ethnopharmacological investigation indicates that the use of these cups is simply due to the pharmacological properties attributed to "bitters", and is strongly linked to tradition and cultural attitudes. This study also emphasizes the long lasting use of these cups, now restricted to Suriname only, with almost no variation over one century.

[Duchamp's ready-made pharmaceutical ampoule]

At the end of 1919, Marcel Duchamp realizes, by means of a Parisian pharmacist, a ready-made that he entitles Air de Paris. We describe in this article the original ampule used by Duchamp.

[Pharmaceutical pottery in the Museum of History of Medicine of Rome "La Sapienza"]

Chemist's and pharmaceutical vases are a historical, aesthetic, cultural, often careless, patrimony that evokes numerous suggestions. It is not possible to distinguish their history from the history of general ceramics in order to reconstruct the context in which these manufactured object were made. Even though manufacture presents a peculiarity of its own it interlaces with the objects of everyday use and with the production designed to any specific and different uses. Ceramics is one of the most flourishing art in Italy; a revival of its techniques arose just in relation with the 'rebirth' of spices commercial trading in Italy.

Safe handling of hazardous drugs

Recommendations for the safe handling of hazardous drugs have been available for more than twenty years. Evidence for continued risk of occupational exposure is abundant; however, nurses' use of the recommended precautions is not universal. This may be related to a lack of information or to a lack of serious concern for the potential hazards. This article includes a discussion of current issues related to handling hazardous drugs in the workplace and a review of the history of safe handling guidelines, current recommendations, and barriers to implementing guidelines in health care settings.

[Biosensors in the pharmaceutical domain]

Biosensors are analytical devices which incorporate a biological component (enzyme, antibody, animal or plant cell, DNA fragments, lipids.) intimately connected to a physical transducer (electrode, optical fibre, vibrating quartz.). This dual configuration allows the study of a great variety of compounds of pharmaceutical interest which react with the biocomponent. The latter is selected depending on the application and the performance criteria requested. Biosensors are suitable for real time monitoring such as in bioreactors, and for the determination of various physiological and pharmacological parameters. Biosensors may be employed in home testing (glucose, lactate.), in hospitals (bedside testing, emergency, surgery, dialysis monitoring, etc.) in clinical laboratory analyses (immunoassays, DNA analysis.) and at research centres. Ideally, a biosensor should be easy to use, allowing direct analysis without sample pre-treatment. Measurements should be automatized and remote controlled. The biosensor may be miniaturized for single use or for implementation in sensor arrays. Applications to microenvironments (in vivo, single cell.) or discrete one shot decentralized tests may also considered.

Antiques of the pharmacy. Part 2: Offering relief when cure is impossible

Images

Instrumental color measurement: a method for judging the appearance of tablets

An instrumental method for the quantification of the color and the discoloration of tablets is presented. The results are not influenced by external factors and are more rapidly accessible compared with visual determination. Two of the measured parameters, lightness (delta L*) and yellowishness (delta b*) were found to reflect the changes in coloration observed in stability tests on white metoprolol tablets. Correlation studies between visual judgments and instrumental measurements are presented. The instrumental performance was carefully examined, and the results indicate that the employed instrument gives color parameters in the same range for tablet samples irrespective of whether the examined tablet surface has an engraving on it or not.

The Giustiniani medicine chest

Images

Retail pharmacy activities and their automation by bar code recorder, tablet counter and remote computer. A feasibility trial

The advent of mechanical tablet counters in retail pharmacy, and the development of methods of stock control involving mechanical code reading in other forms of retailing have prompted the trail of a machine which combines both these functions with support from a computer. The wide range of activities over which the combined machine can provide assistance to the pharmacist has been explored. An extension of the system utilising the same data would provide the basis for automation of the pharmaceutical wholesaler and manufacturer, and of the prescription pricing bureau.

Apparatus for studying the disintegration of tablet coatings

A new apparatus has been designed to assess rates of disintegration of tablet coatings and the availability of coated tablet contents. Random movement of the coated tablet in a known volume of mildly turbulent disintegrating fluid at a standard temperature is achieved in a perforated Perspex drum made to revolve slowly in the fluid which moves up and down in its container at the same time. The procedure eliminates the tendency for coatings to stick together or to the grid during a test, which is seen with present pharmacopoeial designs and which causes large variability in the results.

Apparatus for accelerated temperature cycling

An automatic control unit, activating electrically operated solenoid valves, is used to regulate the flow of “hot” and “cold” water from two thermostated baths to the jacket of a reaction vessel so that the temperature of a liquid sample within the vessel follows a highly reproducible programmed cycle. A comparison is made of the temperature hysteresis between 23–33° during a 16 min cycle, when the flow rates to the vessel are changed and when the valves are operated manually. The adjustment of temperature hysteresis to effect the accelerated storage testing of liquid preparations is described.