Transparent Exploration of Machine Learning for Biomarker Discovery from Proteomics and Omics Data.

Biomarkers are of central importance for assessing the health state and to guide medical interventions and their efficacy; still, they are lacking for most diseases. Mass spectrometry (MS)-based proteomics is a powerful technology for biomarker discovery but requires sophisticated bioinformatics to identify robust patterns. Machine learning (ML) has become a promising tool for this purpose. However, it is sometimes applied in an opaque manner and generally requires specialized knowledge. To enable easy access to ML for biomarker discovery without any programming or bioinformatics skills, we developed "OmicLearn" (http://OmicLearn.org), an open-source browser-based ML tool using the latest advances in the Python ML ecosystem. Data matrices from omics experiments are easily uploaded to an online or a locally installed web server. OmicLearn enables rapid exploration of the suitability of various ML algorithms for the experimental data sets. It fosters open science via transparent assessment of state-of-the-art algorithms in a standardized format for proteomics and other omics sciences.

Hyperspectral Data and Machine Learning for Estimating CDOM, Chlorophyll a, Diatoms, Green Algae and Turbidity.

Inland waters are of great importance for scientists as well as authorities since they are essential ecosystems and well known for their biodiversity. When monitoring their respective water quality, in situ measurements of water quality parameters are spatially limited, costly and time-consuming. In this paper, we propose a combination of hyperspectral data and machine learning methods to estimate and therefore to monitor different parameters for water quality. In contrast to commonly-applied techniques such as band ratios, this approach is data-driven and does not rely on any domain knowledge. We focus on CDOM, chlorophyll a and turbidity as well as the concentrations of the two algae types, diatoms and green algae. In order to investigate the potential of our proposal, we rely on measured data, which we sampled with three different sensors on the river Elbe in Germany from 24 June⁻12 July 2017. The measurement setup with two probe sensors and a hyperspectral sensor is described in detail. To estimate the five mentioned variables, we present an appropriate regression framework involving ten machine learning models and two preprocessing methods. This allows the regression performance of each model and variable to be evaluated. The best performing model for each variable results in a coefficient of determination R 2 in the range of 89.9% to 94.6%. That clearly reveals the potential of the machine learning approaches with hyperspectral data. In further investigations, we focus on the generalization of the regression framework to prepare its application to different types of inland waters.