Convolutional Neural Networks Guided Raman Spectroscopy as a Process Analytical Technology (PAT) Tool for Monitoring and Simultaneous Prediction of Monoclonal Antibody Charge Variants

BACKGROUND

Charge related heterogeneities of monoclonal antibody (mAb) based therapeutic products are increasingly being considered as a critical quality attribute (CQA). They are typically estimated using analytical cation exchange chromatography (CEX), which is time consuming and not suitable for real time control. Raman spectroscopy coupled with artificial intelligence (AI) tools offers an opportunity for real time monitoring and control of charge variants.

OBJECTIVE

We present a process analytical technology (PAT) tool for on-line and real-time charge variant determination during process scale CEX based on Raman spectroscopy employing machine learning techniques.

METHOD

Raman spectra are collected from a reference library of samples with distribution of acidic, main, and basic species from 0-100% in a mAb concentration range of 0-20 g/L generated from process-scale CEX. The performance of different machine learning techniques for spectral processing is compared for predicting different charge variant species.

RESULT

A convolutional neural network (CNN) based model was successfully calibrated for quantification of acidic species, main species, basic species, and total protein concentration with R2 values of 0.94, 0.99, 0.96 and 0.99, respectively, and the Root Mean Squared Error (RMSE) of 0.1846, 0.1627, and 0.1029 g/L, respectively, and 0.2483 g/L for the total protein concentration.

CONCLUSION

We demonstrate that Raman spectroscopy combined with AI-ML frameworks can deliver rapid and accurate determination of product related impurities. This approach can be used for real time CEX pooling decisions in mAb production processes, thus enabling consistent charge variant profiles to be achieved.

Trait anxiety modulates the detection sensitivity of negative affect in speech: an online pilot study

Acoustic perception of emotions in speech is relevant for humans to navigate the social environment optimally. While sensory perception is known to be influenced by ambient noise, and bodily internal states (e.g., emotional arousal and anxiety), their relationship to human auditory perception is relatively less understood. In a supervised, online pilot experiment sans the artificially controlled laboratory environment, we asked if the detection sensitivity of emotions conveyed by human speech-in-noise (acoustic signals) varies between individuals with relatively lower and higher levels of subclinical trait-anxiety, respectively. In a task, participants (n = 28) accurately discriminated the target emotion conveyed by the temporally unpredictable acoustic signals (signal to noise ratio = 10 dB), which were manipulated at four levels (Happy, Neutral, Fear, and Disgust). We calculated the empirical area under the curve (a measure of acoustic signal detection sensitivity) based on signal detection theory to answer our questions. A subset of individuals with High trait-anxiety relative to Low in the above sample showed significantly lower detection sensitivities to acoustic signals of negative emotions - Disgust and Fear and significantly lower detection sensitivities to acoustic signals when averaged across all emotions. The results from this pilot study with a small but statistically relevant sample size suggest that trait-anxiety levels influence the overall acoustic detection of speech-in-noise, especially those conveying threatening/negative affect. The findings are relevant for future research on acoustic perception anomalies underlying affective traits and disorders.

Environmental insults and compensative responses: when microbiome meets cancer

Tumor microenvironment has recently been ascribed a new hallmark-the polymorphic microbiome. Accumulating evidence regarding the tissue specific territories of tumor-microbiome have opened new and interesting avenues. A pertinent question is regarding the functional consequence of the interface between host-microbiome and cancer. Given microbial communities have predominantly been explored through an ecological perspective, it is important that the foundational aspects of ecological stress and the fight to 'survive and thrive' are accounted for tumor-micro(b)environment as well. Building on existing evidence and classical microbial ecology, here we attempt to characterize the ecological stresses and the compensative responses of the microorganisms inside the tumor microenvironment. What insults would microbes experience inside the cancer jungle? How would they respond to these insults? How the interplay of stress and microbial quest for survival would influence the fate of tumor? This work asks these questions and tries to describe this underdiscussed ecological interface of the tumor and its microbiota. It is hoped that a larger scientific thought on the importance of microbial competition sensing vis-à-vis tumor-microenvironment would be stimulated.

Elucidating collective translocation of nanoparticles across the skin lipid matrix: a molecular dynamics study

The top layer of skin, the stratum corneum, provides a formidable barrier to the skin. Nanoparticles are utilized and further explored for personal and health care applications related to the skin. In the past few years, several researchers have studied the translocation and permeation of nanoparticles of various shapes, sizes, and surface chemistry through cell membranes. Most of these studies focused on a single nanoparticle and a simple bilayer system, whereas skin has a highly complex lipid membrane architecture. Moreover, it is highly unlikely that a nanoparticle formulation applied on the skin will not have multiple nanoparticle-nanoparticle and skin-nanoparticle interactions. In this study, we have utilized coarse-grained MARTINI molecular dynamics simulations to assess the interactions of two types (bare and dodecane-thiol coated) of nanoparticles with two models (single bilayer and double bilayer) of skin lipid membranes. The nanoparticles were found to be partitioned from the water layer to the lipid membrane as an individual entity as well as in the cluster form. It was discovered that each nanoparticle reached the interior of both single bilayer and double bilayer membranes irrespective of the nanoparticle type and concentration, though coated particles were observed to efficiently traverse across the bilayer when compared with bare particles. The coated nanoparticles also created a single large cluster inside the membrane, whereas the bare nanoparticles were found in small clusters. Both the nanoparticles exhibited preferential interactions with cholesterol molecules present in the lipid membrane as compared to other lipid components of the membrane. We have also observed that the single membrane model exhibited unrealistic instability at moderate to higher concentrations of nanoparticles, and hence for translocation study, a minimum double bilayer model should be employed.

When less is more powerful: Shapley value attributed ablation with augmented learning for practical time series sensor data classification

Time series sensor data classification tasks often suffer from training data scarcity issue due to the expenses associated with the expert-intervened annotation efforts. For example, Electrocardiogram (ECG) data classification for cardio-vascular disease (CVD) detection requires expensive labeling procedures with the help of cardiologists. Current state-of-the-art algorithms like deep learning models have shown outstanding performance under the general requirement of availability of large set of training examples. In this paper, we propose Shapley Attributed Ablation with Augmented Learning: ShapAAL, which demonstrates that deep learning algorithm with suitably selected subset of the seen examples or ablating the unimportant ones from the given limited training dataset can ensure consistently better classification performance under augmented training. In ShapAAL, additive perturbed training augments the input space to compensate the scarcity in training examples using Residual Network (ResNet) architecture through perturbation-induced inputs, while Shapley attribution seeks the subset from the augmented training space for better learnability with the goal of better general predictive performance, thanks to the "efficiency" and "null player" axioms of transferable utility games upon which Shapley value game is formulated. In ShapAAL, the subset of training examples that contribute positively to a supervised learning setup is derived from the notion of coalition games using Shapley values associated with each of the given inputs' contribution into the model prediction. ShapAAL is a novel push-pull deep architecture where the subset selection through Shapley value attribution pushes the model to lower dimension while augmented training augments the learning capability of the model over unseen data. We perform ablation study to provide the empirical evidence of our claim and we show that proposed ShapAAL method consistently outperforms the current baselines and state-of-the-art algorithms for time series sensor data classification tasks from publicly available UCR time series archive that includes different practical important problems like detection of CVDs from ECG data.

Peptide mediated colorimetric detection of SARS-CoV-2 using gold nanoparticles: a molecular dynamics simulation study

The outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has necessitated the development of a rapid, simple yet selective naked-eye detection methodology that does not require any advanced instrumental techniques. In this study, we report our computational findings on the detection of SARS-CoV-2 using peptide- functionalized gold nanoparticles (GNPs). The peptide has been screened from angiotensin-converting enzyme 2 (ACE2) receptor situated on the surface of the host cell membrane which interacts with the spike protein of SARS-CoV-2, resulting entry of the virus into the host cell. As a result, the peptide-functionalized GNPs possess excellent affinity towards the spikes of SARS-CoV-2 and readily get aggregated once exposed to SARS-CoV-2 antigen or virus. The stability of the peptides on the surface of GNPs and their interaction with the spike protein of the virus have been investigated using coarse-grained molecular dynamic simulations. The potential of mean force calculation of spike protein confirmed strong binding between peptide and receptor-binding domain (RBD) of spike protein. Our in silico results demonstrate the potential of the peptide-functionalized GNPs in the development of simple and rapid colorimetric biosensors for clinical diagnosis.

CurmElo: The theory and practice of a forced-choice approach to producing preference rankings

We introduce CurmElo, a forced-choice approach to producing a preference ranking of an arbitrary set of objects that combines the Elo algorithm with novel techniques for detecting and correcting for (1) preference heterogeneity induced polarization in preferences among raters, and (2) intransitivity in preference rankings. We detail the application of CurmElo to the problem of generating approximately preference-neutral identifiers, in this case four-letter and five-letter nonsense words patterned on the phonological conventions of the English language, using a population of Amazon Mechanical Turk workers. We find evidence that human raters have significant non-uniform preferences over these nonsense words, and we detail the consequences of this finding for social science work that utilizes identifiers without accounting for the bias this can induce. In addition, we describe how CurmElo can be used to produce rankings of arbitrary features or dimensions of preference of a set of objects relative to a population of raters.

Multimodal cardiovascular model for hemodynamic analysis: Simulation study on mitral valve disorders

Valvular heart diseases are a prevalent cause of cardiovascular morbidity and mortality worldwide, affecting a wide spectrum of the population. In-silico modeling of the cardiovascular system has recently gained recognition as a useful tool in cardiovascular research and clinical applications. Here, we present an in-silico cardiac computational model to analyze the effect and severity of valvular disease on general hemodynamic parameters. We propose a multimodal and multiscale cardiovascular model to simulate and understand the progression of valvular disease associated with the mitral valve. The developed model integrates cardiac electrophysiology with hemodynamic modeling, thus giving a broader and holistic understanding of the effect of disease progression on various parameters like ejection fraction, cardiac output, blood pressure, etc., to assess the severity of mitral valve disorders, naming Mitral Stenosis and Mitral Regurgitation. The model mimics an adult cardiovascular system, comprising a four-chambered heart with systemic, pulmonic circulation. The simulation of the model output comprises regulated pressure, volume, and flow for each heart chamber, valve dynamics, and Photoplethysmogram signal for normal physiological as well as pathological conditions due to mitral valve disorders. The generated physiological parameters are in agreement with published data. Additionally, we have related the simulated left atrium and ventricle dimensions, with the enlargement and hypertrophy in the cardiac chambers of patients with mitral valve disorders, using their Electrocardiogram available in Physionet PTBI dataset. The model also helps to create 'what if' scenarios and relevant analysis to study the effect in different hemodynamic parameters for stress or exercise like conditions.

Structural aspects of lesional and non-lesional skin microbiota reveal key community changes in leprosy patients from India

Although skin is the primary affected organ in Leprosy, the role of the skin microbiome in its pathogenesis is not well understood. Recent reports have shown that skin of leprosy patients (LP) harbours perturbed microbiota which grants inflammation and disease progression. Herein, we present the results of nested Polymerase Chain Reaction-Denaturing Gradient Gel Electrophoresis (PCR-DGGE) which was initially performed for investigating the diversity of bacterial communities from lesional skin (LS) and non-lesional skin (NLS) sites of LP (n = 11). Further, we performed comprehensive analysis of 16S rRNA profiles corresponding to skin samples from participants (n = 90) located in two geographical locations i.e. Hyderabad and Miraj in India. The genus Staphylococcus was observed to be one of the representative bacteria characterizing healthy controls (HC; n = 30), which in contrast was underrepresented in skin microbiota of LP. Taxa affiliated to phyla Firmicutes and Proteobacteria were found to be signatures of HC and LS, respectively. Observed diversity level changes, shifts in core microbiota, and community network structure support the evident dysbiosis in normal skin microbiota due to leprosy. Insights obtained indicate the need for exploring skin microbiota modulation as a potential therapeutic option for leprosy.

Applied machine learning for predicting the lanthanide-ligand binding affinities

Binding affinities of metal-ligand complexes are central to a multitude of applications like drug design, chelation therapy, designing reagents for solvent extraction etc. While state-of-the-art molecular modelling approaches are usually employed to gather structural and chemical insights about the metal complexation with ligands, their computational cost and the limited ability to predict metal-ligand stability constants with reasonable accuracy, renders them impractical to screen large chemical spaces. In this context, leveraging vast amounts of experimental data to learn the metal-binding affinities of ligands becomes a promising alternative. Here, we develop a machine learning framework for predicting binding affinities (logK1) of lanthanide cations with several structurally diverse molecular ligands. Six supervised machine learning algorithms-Random Forest (RF), k-Nearest Neighbours (KNN), Support Vector Machines (SVM), Kernel Ridge Regression (KRR), Multi Layered Perceptrons (MLP) and Adaptive Boosting (AdaBoost)-were trained on a dataset comprising thousands of experimental values of logK1 and validated in an external 10-folds cross-validation procedure. This was followed by a thorough feature engineering and feature importance analysis to identify the molecular, metallic and solvent features most relevant to binding affinity prediction, along with an evaluation of performance metrics against the dimensionality of feature space. Having demonstrated the excellent predictive ability of our framework, we utilized the best performing AdaBoost model to predict the logK1 values of lanthanide cations with nearly 71 million compounds present in the PubChem database. Our methodology opens up an opportunity for significantly accelerating screening and design of ligands for various targeted applications, from vast chemical spaces.

Weaving nanostructures with site-specific ion induced bidirectional bending

Site-specific ion-irradiation is a promising tool fostering strain-engineering of freestanding nanostructures to realize 3D-configurations towards various functionalities. We first develop a novel approach of fabricating freestanding 3D silicon nanostructures by low dose ion-implantation followed by chemical-etching. The fabricated nanostructures can then be deformed bidirectionally by varying the local irradiation of kiloelectronvolt gallium ions. By further tuning the ion-dose and energy, various nanostructure configurations can be realized, thus extending its horizon to new functional 3D-nanostructures. It has been revealed that at higher-energies (∼30 kV), the nanostructures can exhibit two-stage bidirectional-bending in contrast to the bending towards the incident-ions at lower-energies (∼16), implying an effective transfer of kinetic-energy. Computational studies show that the spatial-distribution of implanted-ions, dislocated silicon atoms, has potentially contributed to the local development of stresses. Nanocharacterization confirms the formation of two distinguishable ion-irradiated and un-irradiated regions, while the smoothened morphology of the irradiated-surface suggested that the bending is also coupled with sputtering at higher ion-doses. The bending effects associated with local ion irradiation in contrast to global ion irradiation are presented, with the mechanism elucidated. Finally, weaving of nanostructures is demonstrated through strain-engineering for new nanoscale artefacts such as ultra-long fully-bent nanowires, nano-hooks, and nano-meshes. The aligned growth of bacterial-cells is observed on the fabricated nanowires, and a mesh based "bacterial-trap" for site-specific capture of bacterial cells is demonstrated emphasizing the versatile nature of the current approach.