Advancing computational biology and bioinformatics research through open innovation competitions

Responsiveness of open innovation to COVID-19 pandemic: The case of data for good

Due to the COVID-19 pandemic, countries around the world are facing one of the most severe health and economic crises of recent history and human society is called to figure out effective responses. However, as current measures have not produced valuable solutions, a multidisciplinary and open approach, enabling collaborations across private and public organizations, is crucial to unleash successful contributions against the disease. Indeed, the COVID-19 represents a Grand Challenge to which joint forces and extension of disciplinary boundaries have been recognized as main imperatives. As a consequence, Open Innovation represents a promising solution to provide a fast recovery. In this paper we present a practical application of this approach, showing how knowledge sharing constitutes one of the main drivers to tackle pressing social needs. To demonstrate this, we propose a case study regarding a data sharing initiative promoted by Facebook, the Data For Good program. We leverage a large-scale dataset provided by Facebook to the research community to offer a representation of the evolution of the Italian mobility during the lockdown. We show that this repository allows to capture different patterns of movements on the territory with increasing levels of detail. We integrate this information with Open Data provided by the Lombardy region to illustrate how data sharing can also provide insights for private businesses and local authorities. Finally, we show how to interpret Data For Good initiatives in light of the Open Innovation Framework and discuss the barriers to adoption faced by public administrations regarding these practices.

Improving deconvolution methods in biology through open innovation competitions: an application to the connectivity map

MOTIVATION

Do machine learning methods improve standard deconvolution techniques for gene expression data? This article uses a unique new dataset combined with an open innovation competition to evaluate a wide range of approaches developed by 294 competitors from 20 countries. The competition's objective was to address a deconvolution problem critical to analyzing genetic perturbations from the Connectivity Map. The issue consists of separating gene expression of individual genes from raw measurements obtained from gene pairs. We evaluated the outcomes using ground-truth data (direct measurements for single genes) obtained from the same samples.

RESULTS

We find that the top-ranked algorithm, based on random forest regression, beat the other methods in accuracy and reproducibility; more traditional gaussian-mixture methods performed well and tended to be faster, and the best deep learning approach yielded outcomes slightly inferior to the above methods. We anticipate researchers in the field will find the dataset and algorithms developed in this study to be a powerful research tool for benchmarking their deconvolution methods and a resource useful for multiple applications.

AVAILABILITY AND IMPLEMENTATION

The data is freely available at clue.io/data (section Contests) and the software is on GitHub at https://github.com/cmap/gene_deconvolution_challenge.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Rapid crowdsourced innovation for COVID-19 response and economic growth

The COVID-19 pandemic has profoundly affected life worldwide. Governments have been faced with the formidable task of implementing public health measures, such as social distancing, quarantines, and lockdowns, while simultaneously supporting a sluggish economy and stimulating research and development (R&D) for the pandemic. Catalyzing bottom-up entrepreneurship is one method to achieve this. Home-grown efforts by citizens wishing to contribute their time and resources to help have sprouted organically, with ideas shared widely on the internet. We outline a framework for structured, crowdsourced innovation that facilitates collaboration to tackle real, contextualized problems. This is exemplified by a series of virtual hackathon events attracting over 9000 applicants from 142 countries and 49 states. A hackathon is an event that convenes diverse individuals to crowdsource solutions around a core set of predetermined challenges in a limited amount of time. A consortium of over 100 partners from across the healthcare spectrum and beyond defined challenges and supported teams after the event, resulting in the continuation of at least 25% of all teams post-event. Grassroots entrepreneurship can stimulate economic growth while contributing to broader R&D efforts to confront public health emergencies.

Exploring the Use of Compound-Induced Transcriptomic Data Generated From Cell Lines to Predict Compound Activity Toward Molecular Targets

Pharmaceutical or phytopharmaceutical molecules rely on the interaction with one or more specific molecular targets to induce their anticipated biological responses. Nonetheless, these compounds are also prone to interact with many other non-intended biological targets, also known as off-targets. Unfortunately, off-target identification is difficult and expensive. Consequently, QSAR models predicting the activity on a target have gained importance in drug discovery or in the de-risking of chemicals. However, a restricted number of targets are well characterized and hold enough data to build such in silico models. A good alternative to individual target evaluations is to use integrative evaluations such as transcriptomics obtained from compound-induced gene expression measurements derived from cell cultures. The advantage of these particular experiments is to capture the consequences of the interaction of compounds on many possible molecular targets and biological pathways, without having any constraints concerning the chemical space. In this work, we assessed the value of a large public dataset of compound-induced transcriptomic data, to predict compound activity on a selection of 69 molecular targets. We compared such descriptors with other QSAR descriptors, namely the Morgan fingerprints (similar to extended-connectivity fingerprints). Depending on the target, active compounds could show similar signatures in one or multiple cell lines, whether these active compounds shared similar or different chemical structures. Random forest models using gene expression signatures were able to perform similarly or better than counterpart models built with Morgan fingerprints for 25% of the target prediction tasks. These performances occurred mostly using signatures produced in cell lines showing similar signatures for active compounds toward the considered target. We show that compound-induced transcriptomic data could represent a great opportunity for target prediction, allowing to overcome the chemical space limitation of QSAR models.