Chronic pulmonary MRSA disease in cystic fibrosis (CF) has a high probably of recurrence following treatment with standard-of-care antibiotics and represents an area of unmet need associated with reduced life expectancy. We developed a lipoglycopeptide therapy customized for pulmonary delivery that not only demonstrates potent activity against planktonic MRSA but also against protected colonies of MRSA both in biofilms and within cells, the latter of which have been linked to clinical antibiotic failure. A library of next-generation potent lipoglycopeptides were synthesized with an emphasis on attaining superior pharmacokinetics (PK) and pharmacodynamics to similar compounds of their class. Our strategy focused on hydrophobic modification of vancomycin where ester and amide functionality were included with carbonyl configuration and alkyl length as key variables. Candidates representative of each carbonyl attachment chemistry demonstrated potent activity in vitro with several compounds being 30-60 times more potent than vancomycin. Several compounds were advanced into in vivo nose-only inhalation PK evaluations in rats where RV94, a potent lipoglycopeptide that utilizes an inverted amide linker to attach a 10-carbon chain to vancomycin, demonstrated the most favorable lung residence time after inhalation. Further in vitro evaluation of RV94 showed superior activity to vancomycin against an expanded panel of gram-positive organisms, cellular accumulation and efficacy against intracellular MRSA, and MRSA biofilm killing. Moreover, in vivo efficacy of inhaled nebulized RV94 in a 48-h acute model of pulmonary MRSA (USA300) infection in neutropenic rats demonstrated statistically significant antibacterial activity that was superior to inhaled vancomycin.

No abstract

A powerful way to improve performance in machine learning is to construct an ensemble that combines the predictions of multiple models. Ensemble methods are often much more accurate and lower variance than the individual classifiers that make them up, but have high requirements in terms of memory and computational time. In fact, a large number of alternative algorithms is usually adopted, each requiring to query all available data. We propose a new quantum algorithm that exploits quantum superposition, entanglement and interference to build an ensemble of classification models. Thanks to the generation of the several quantum trajectories in superposition, we obtain B transformations of the quantum state which encodes the training set in only log (B) operations. This implies an exponential speed-up in the size of the ensemble with respect to classical methods. Furthermore, when considering the overall cost of the algorithm, we show that the training of a single weak classifier impacts additively rather than multiplicatively, as it usually happens. We also present small-scale experiments, defining a quantum version of the cosine classifier and using the IBM qiskit environment to show how the algorithm works.

No abstract

A powerful way to improve performance in machine learning is to construct an ensemble that combines the predictions of multiple models. Ensemble methods are often much more accurate and have a lower variance than the individual classifiers that make them up, but have high requirements in terms of memory and computational time. In fact, a large number of alternative algorithms is usually adopted, each requiring to query all available data. We propose a new quantum algorithm that exploits quantum superposition, entanglement and interference to build an ensemble of classification models. Thanks to the generation of the several quantum trajectories in superposition, we obtain B transformations of the quantum state which encodes the training set in only log ( B ) operations. This implies an exponential growth of the ensemble size while increasing linearly the depth of the correspondent circuit. Furthermore, when considering the overall cost of the algorithm, we show that the training of a single weak classifier impacts additively to the overall time complexity rather than multiplicatively, as it usually happens in classical ensemble methods. We also present small-scale experiments on real-world datasets, defining a quantum version of the cosine classifier and using the IBM qiskit environment to show how the algorithms work.