PlasmoFAB: a benchmark to foster machine learning for Plasmodium falciparum protein antigen candidate prediction.

MOTIVATION: Machine learning methods can be used to support scientific discovery in healthcare-related research fields. However, these methods can only be reliably used if they can be trained on high-quality and curated datasets. Currently, no such dataset for the exploration of Plasmodium falciparum protein antigen candidates exists. The parasite P.falciparum causes the infectious disease malaria. Thus, identifying potential antigens is of utmost importance for the development of antimalarial drugs and vaccines. Since exploring antigen candidates experimentally is an expensive and time-consuming process, applying machine learning methods to support this process has the potential to accelerate the development of drugs and vaccines, which are needed for fighting and controlling malaria. RESULTS: We developed PlasmoFAB, a curated benchmark that can be used to train machine learning methods for the exploration of P.falciparum protein antigen candidates. We combined an extensive literature search with domain expertise to create high-quality labels for P.falciparum specific proteins that distinguish between antigen candidates and intracellular proteins. Additionally, we used our benchmark to compare different well-known prediction models and available protein localization prediction services on the task of identifying protein antigen candidates. We show that available general-purpose services are unable to provide sufficient performance on identifying protein antigen candidates and are outperformed by our models that were trained on this tailored data. AVAILABILITY AND IMPLEMENTATION: PlasmoFAB is publicly available on Zenodo with DOI 10.5281/zenodo.7433087. Furthermore, all scripts that were used in the creation of PlasmoFAB and the training and evaluation of machine learning models are open source and publicly available on GitHub here: https://github.com/msmdev/PlasmoFAB.

Marker-free phenotyping of tumor cells by fractal analysis of reflection interference contrast microscopy images.

Phenotyping of tumor cells by marker-free quantification is important for cancer diagnostics. For the first time, fractal analysis of reflection interference contrast microscopy images of single living cells was employed as a new method to distinguish between different nanoscopic membrane features of tumor cells. Since tumor progression correlates with a higher degree of chaos within the cell, it can be quantified mathematically by fractality. Our results show a high accuracy in identifying malignant cells with a failure chance of 3%, which is far better than today's applied methods.

Flow conditions in the vicinity of microstructured interfaces studied by holography and implications for the assembly of artificial actin networks.

Microstructured fluidic devices have successfully been used for the assembly of free standing actin networks as mechanical model systems on the top of micropillars. The assembly occurs spontaneously at the pillar heads when preformed filaments are injected into the channel. In order to reveal the driving mechanism of this localization, we studied the properties of the flow profile by holographic tracking. Despite the strong optical disturbances originating from the pillar field, 2 µm particles were traced with digital in-line holographic microscopy (DIHM). Trajectories in the pillar free region and local alterations of the flow profile induced by the channel structure in the pillar decorated region can be distinguished. Velocity histograms at different z-positions reveal that the laminar flow profile across the channel shows a difference between the minimum in the z-component of the velocity field and the maximum of the overall velocity. This minimum drag in vertical direction is present at the top of the pillars and explains why biopolymer networks readily assemble in this region instead of forming a homogeneous three-dimensional network in between the pillars. On the basis of the observations we propose a new mechanism for actin network formation on top of the microstructures.

Motion Biomarkers Showing Maximum Contrast Between Healthy Subjects and Parkinson's Disease Patients Treated With Deep Brain Stimulation of the Subthalamic Nucleus. A Pilot Study.

Background: Classic motion abnormalities in Parkinson's disease (PD), such as tremor, bradykinesia, or rigidity, are well-covered by standard clinical assessments such as the Unified Parkinson's Disease Rating Scale (UPDRS). However, PD includes motor abnormalities beyond the symptoms and signs as measured by UPDRS, such as the lack of anticipatory adjustments or compromised movement smoothness, which are difficult to assess clinically. Moreover, PD may entail motor abnormalities not yet known. All these abnormalities are quantifiable via motion capture and may serve as biomarkers to diagnose and monitor PD. Objective: In this pilot study, we attempted to identify motion features revealing maximum contrast between healthy subjects and PD patients with deep brain stimulation (DBS) of the nucleus subthalamicus (STN) switched off and on as the first step to develop biomarkers for detecting and monitoring PD patients' motor symptoms. Methods: We performed 3D gait analysis in 7 out of 26 PD patients with DBS switched off and on, and in 25 healthy control subjects. We computed feature values for each stride, related to 22 body segments, four time derivatives, left-right mean vs. difference, and mean vs. variance across stride time. We then ranked the feature values according to their distinguishing power between PD patients and healthy subjects. Results: The foot and lower leg segments proved better in classifying motor anomalies than any other segment. Higher degrees of time derivatives were superior to lower degrees (jerk > acceleration > velocity > displacement). The averaged movements across left and right demonstrated greater distinguishing power than left-right asymmetries. The variability of motion was superior to motion's absolute values. Conclusions: This small pilot study identified the variability of a smoothness measure, i.e., jerk of the foot, as the optimal signal to separate healthy subjects' from PD patients' gait. This biomarker is invisible to clinicians' naked eye and is therefore not included in current motor assessments such as the UPDRS. We therefore recommend that more extensive investigations be conducted to identify the most powerful biomarkers to characterize motor abnormalities in PD. Future studies may challenge the composition of traditional assessments such as the UPDRS.

Reversibility and Viscoelastic Properties of Micropillar Supported and Oriented Magnesium Bundled F-Actin.

Filamentous actin is one of the most important cytoskeletal elements. Not only is it responsible for the elastic properties of many cell types, but it also plays a vital role in cellular adhesion and motility. Understanding the bundling kinetics of actin filaments is important in the formation of various cytoskeletal structures, such as filopodia and stress fibers. Utilizing a unique pillar-structured microfluidic device, we investigated the time dependence of bundling kinetics of pillar supported free-standing actin filaments. Microparticles attached to the filaments allowed the measurement of thermal motion, and we found that bundling takes place at lower concentrations than previously found in 3-dimensional actin gels, i.e. actin filaments formed bundles in the presence of 5-12 mM of magnesium chloride in a time-dependent manner. The filaments also displayed long term stability for up to hours after removing the magnesium ions from the buffer, which suggests that there is an extensive hysteresis between cation induced crosslinking and decrosslinking.

Python algorithms in particle tracking microrheology.

BACKGROUND: Particle tracking passive microrheology relates recorded trajectories of microbeads, embedded in soft samples, to the local mechanical properties of the sample. The method requires intensive numerical data processing and tools allowing control of the calculation errors. RESULTS: We report the development of a software package collecting functions and scripts written in Python for automated and manual data processing, to extract viscoelastic information about the sample using recorded particle trajectories. The resulting program package analyzes the fundamental diffusion characteristics of particle trajectories and calculates the frequency dependent complex shear modulus using methods published in the literature. In order to increase conversion accuracy, segmentwise, double step, range-adaptive fitting and dynamic sampling algorithms are introduced to interpolate the data in a splinelike manner. CONCLUSIONS: The presented set of algorithms allows for flexible data processing for particle tracking microrheology. The package presents improved algorithms for mean square displacement estimation, controlling effects of frame loss during recording, and a novel numerical conversion method using segmentwise interpolation, decreasing the conversion error from about 100% to the order of 1%.

Splinelike interpolation in particle tracking microrheology.

Converting time dependent creep compliance to frequency dependent complex shear modulus is an important step in analyzing the results of particle tracking microrheology. Fitting a function to the whole time range and transforming it to calculate the shear modulus is one way of solving this problem. However, the creep compliance of many samples, such as gels of biopolymers, shows different trends under different time regimes. Fitting in these regimes segmentwise results in a function which usually cannot be transformed in a closed analytical form. In general, unlike for beta and cubic splines, also the continuity of the first derivative cannot be ensured. In this paper, we present a method for using segmentwise fitting and numerical conversion, discussing interpolation for improving the transition between the fitted ranges, and propose a dynamic sampling technique to control the accuracy of the resultant complex shear modulus.

A quantitative 3D motility analysis of Trypanosoma brucei by use of digital in-line holographic microscopy.

We present a quantitative 3D analysis of the motility of the blood parasite Trypanosoma brucei. Digital in-line holographic microscopy has been used to track single cells with high temporal and spatial accuracy to obtain quantitative data on their behavior. Comparing bloodstream form and insect form trypanosomes as well as mutant and wildtype cells under varying external conditions we were able to derive a general two-state-run-and-tumble-model for trypanosome motility. Differences in the motility of distinct strains indicate that adaption of the trypanosomes to their natural environments involves a change in their mode of swimming.