R. Canonico

A principal components algorithm for spectra normalisation

There is an increasing use of spectroscopic techniques, such as high-resolution NMR spectroscopy, to examine variations in cell metabolism and/or structure in response to numerous physical, chemical, and biological agents. In these types of studies, in order to obtain relative quantitative information, a comparison between signal intensities of control samples and treated or exposed ones is often conducted. A possible strategy is to estimate, by an opportune algorithm, a normalisation constant which takes into consideration all cell metabolites in the sample. In this paper, a new normalisation algorithm based on Principal Component Analysis (PCA) is presented. PRICONA (PRIncipal COmponent Normalisation Algorithm) is advantageous in normalising simultaneously great data sets of spectra, in individuating signals that could have been affected by the agent, and in allowing to measure their quantitative variation. The algorithm was tested by Monte Carlo simulations as well as experimentally.

Comparison of 1H-NMR spectra by normalisation algorithms for studying amyloid toxicity in cells

A range of debilitating human diseases is associated with the formation of stable, highly organised, protein aggregates, the amyloid fibrils. Substantial evidence suggests that Prefibrillar Oligomeric Aggregates (PFAs), preceding mature fibrils formation, are the crucial species in the onset of the neuronal degeneration even if with mechanisms to be further cleared. In this work, we show how 1H-NMR cell spectral analysis methods can prove to be very effective tools to clear the PFAs amyloid cytotoxicity mechanisms. Following the same method shown by Vilasi, we apply a new 1H-NMR analysis algorithm to identify the metabolites significantly varied in cells incubated with toxic oligomers from the amyloidogenic W7FW14F mutant of apomyoglobin. Our main aim is to confirm the results obtained by Vilasi et al., normalising a set of different data spectra with the new PRICONA algorithm here described, thus contributing to strengthen the general framework of metabolites and proteins involved in cellular amyloid toxicity.

Long term seismic noise acquisition and analysis with tunable monolithic horizontal sensors at the INFN Gran Sasso National Laboratory

In this paper we present the scientific data recorded by tunable mechanical monolithic horizontal seismometers located in the Gran Sasso National Laboratory of the INFN, within thermally insulating enclosures onto concrete slabs connected to the bedrock. The main goals of this long term test are a preliminary seismic characterization of the site in the frequency band 10−7÷1Hz and the acquisition of all the relevant information for the optimization of the sensors.

Biological NMR FIDs and spectra normalization

There is increasing use of spectroscopic techniques, such as high-resolution NMR spectroscopy, to examine variations in cell metabolism and / or structure in response to numerous physical, chemical, and biological agents. In these types of studies, in order to obtain relative quantitative information, a comparison between signal intensities of control samples and treated or exposed ones is often conducted. The methods thus far developed for this purpose are not directly related to the overall intrinsic properties of the samples, but rather to the addition of external substances of known concentrations or to indirect measurement of internal substances. Another possibility is to estimate, by an opportune algorithm, a normalization constant which takes into consideration all cell metabolites present in the sample. Recently, a new normalization algorithm, based on Principal Component Analysis (PCA), was presented. PCA is a well-known statistical technique for analysis of large, multivariate datasets, which extracts the basic features of the data. The PRICONA (PRincipal COmponent Normalization Algorithm) algorithm use PCA in a new totally different manner: PCA is, in fact, used to normalize spectra in order to obtain quantitative information about the treatment effects. In this paper, a comparison of results obtained in the time domain, that is on NMR FIDs (Free Induction Decay) and in the frequency domain, on NMR spectra, is conducted. The comparison is useful because in NMR spectroscopy analysis in the different domain can have different advantages. The algorithm was tested by Monte Carlo simulations of NMR FIDs.

Low frequency/high sensitivity triaxial monolithic sensor

This paper describes a new mechanical implementation of a triaxial sensor, configurable as seismometer and/or as accelerometer, consisting of three one-dimensional monolithic FP sensors, suitably geometrically positioned. The triaxial sensor is, therefore, compact, light, scalable, tunable instrument (frequency < 100 mHz with large band (10−7 Hz − 10 Hz), high quality factor (Q < 1500 in air) with good immunity to environmental noises, guaranteed by an integrated laser optical readout. The measured sensitivity curve is in very good agreement with the theoretical ones (10−12m/√Hz) in the band (0.1 ÷ 10Hz). Typical applications are in the field of earthquake engineering, geophysics, civil engineering and in all applications requiring large band-low frequency performances coupled with high sensitivities.

Low frequency control strategy for seismic attenuation in inertial platforms and mechanical suspensions

This paper describes a new application of the monolithic folded pendulum configured as seismometer (no force feed-back) and used as sensor in the control of inertial platforms and suspensions, like, for example, those used in interferometric detectors of gravitational waves, where a residual horizontal motion better than 10−15m/√Hz in the band 0.01 ÷ 100Hz is a requirement. The experimental results, obtained in the band 0.01 ÷ 10Hz, demonstrate that this sensor has enough dynamics and sensitivity to introduce no limitations to the state-of-the-art control systems. Moreover, its full scalability allows an easy integration and positioning also on the different stages of multistage mechanical suspensions (seismic attenuators) and inertial platforms. This new application demonstrates not only the feasibility of the proposed new control strategy in the low frequency region, but, and it is very relevant, that it is now possible the implementation of very effective control systems with a large reduction of control electronics, replaced by less noisy optical and mechanical devices, with the further advantage of rendering the whole system surely less sensitive to environmental noises. The results of this study, although preliminary and obtained with sensors not optimized for the specific application, are presented and discussed in this paper, in connection with some of the possible applications (platforms and mechanical structure control and stabilization, building controls, etc.) and the planned further developments and improvements.

The PRICONA algorithm for biological spectra normalization

There is increasing use of spectroscopic techniques, such as high-resolution NMR spectroscopy, to examine variations in cell metabolism and/or structure in response to numerous physical, chemical, and biological agents. In these types of studies, in order to obtain relative quantitative information, a comparison between signal intensities of control samples and treated or exposed ones is often conducted. The methods thus far developed for this purpose are not directly related to the overall intrinsic properties of the samples, but rather to the addition of external substances of known concentrations or to indirect measurement of internal substances. Another possibility is to estimate, by an opportune algorithm, a normalization constant which takes into consideration all cell metabolites present in the sample. Recently, a new normalization algorithm, based on Principal Component Analysis (PCA), was presented. PCA is a well-known statistical technique for analysis of large, multivariate datasets, which extracts the basic features of the data. The PRICONA (PRincipal COmponent Normalization Algorithm) algorithm use PCA in a new totally different manner: PCA is, in fact, used to normalize spectra in order to obtain quantitative information about the treatment effects. In this paper, it is shown that PRICONA can be used in the time domain, that is on NMR FIDs (Free Induction Decay) instead of on NMR spectra. That is advantageous because NMR FIDs do not require any operator dependent manipulation. The algorithm was tested by Monte Carlo simulations of NMR FIDs.

Large-band seismic characterization of the INFN Gran Sasso National Laboratory

In this paper we present the scientific data recorded by tunable mechanical monolithic horizontal seismometers located in the Gran Sasso National Laboratory of the INFN, within thermally insulating enclosures onto concrete slabs connected to the bedrock. The main goals of this long-term large-band measurements are for the seismic characterization of the site in the frequency band 10−6÷10Hz and the acquisition of all the relevant information for the optimization of the sensors.

Low frequency/high sensitivity horizontal monolithic sensor

This paper describes a new mechanical implementation of a folded pendulum based inertial sensor, configurable as seismometer and as accelerometer.1 The sensor is compact, light, scalable, tunable (< 100mHz), with large band (10−6 Hz÷10Hz), high quality factor (Q < 1500 in air) instrument and good immunity to environmental noises, guaranteed by an integrated laser optical readout. The measured sensitivity curve is in very good agreement with the theoretical one (10−12 m/ √ Hz in the band (0.1 ÷ 10Hz). Typical applications are in the field of earthquake engineering, geophysics, and in all applications requiring large band-low frequency performances coupled with high sensitivities.

Mechanical monolithic tiltmeter for low frequency measurements

The paper describes a tilt meter sensor for geophysical applications, based on Folded Pendulum (FP) mechanical sensor. Both the theoretical model and the experimental results of a tunable mechanical monolithic FP tilt meter prototype are presented and discussed. Some of the most important characteristics, like the measured resolution of ≈ 0.1 nrad at 100mHz, are detailed. Among the scientific results, earth tilt tides have been already observed with this monolithic FP tilt meter prototype.