Trantrace: a reliable, traceable, and collaborative solution to streamline the translation of emerging biomedical resources

Background Precision medicine is gaining popularity in routine health care, which heavily relies on the interpretation of emerging biomedical databases and professional guidelines. Translation of those biomedical materials, that is crucial to deliver research discovery to health interventions in non-native English-speaking countries, remains a large amount of work for biomedical practitioners. Results We presented an application called Trantrace to facilitate the routine of large-scale long-running translation. Especially for the case that many people joined multiple translation projects, it has a rigorous task division and cooperation process: assignment, translation, review, release, and iteration. Trantrace empowers users to build own projects with different permissions, version control all the translation operation without specific skills, and further improve the translation efficiency through real-time task management tools. Conclusions By providing a working platform for collaborative translation and enabling long-time version correction, the Trantrace contributes towards the creation of a multilingual biomedical knowledge commons, which would be a valuable aid for personalized therapies. The source code is freely accessible at https://github.com/sgi-drylab/trantrace.


Background
In December 2018, Genomics England accomplished the sequencing 100,000 whole genomes from 70,000 patients and family members, which may serve as a clarion call to integrate genome sequencing into routine medical care [1]. The Cancer Genome Atlas (TCGA), which collected more than 11,000 cases across 33 tumor types, is creating lasting value to elucidate the complex and multi-dimensional problems of human cancers [2], and efforts of harmonized and standardized analysis, such as the Pan-Cancer Atlas, are tried to improve the ability to diagnose, treat and prevent cancers. The Human Microbiome Project (HMP) has been carried out to reveal contributions of the microbiome and its dynamics changes to human health and disease [3]. As more and more national and international genome programs gradually come to an end, it would usher in a new era of personalized medicine and early diagnosis. The rapidly increasing human genotype-phenotype databases collect flooded information about genetic variations and their phenotypic effect based on expert curation of published and unpublished data, which would provide reliable and comprehensive knowledge about the gene and related clinical or research topics [4].
Furthermore, the American College of Medical Genetics (ACMG) and the Association of Molecular Pathologists (AMP) had published a joint consensus to standardize the interpretation of sequence variants [5]. The Clinical Pharmacogenetics Implementation Consortium (CPIC) has published a series of evidence-based guidelines to individualize the prescribing of drugs [6]. Translation of these biomedical resources is crucial to tackle language barriers and bring the latest clinical and molecular findings or research subjects to everyone's health care.
Traditional file-based working with Microsoft may have many pitfalls, including auto-formatting, not traceable and permission problems. A screen of supplementary Excel files in 2016 reported that one in five genetic papers in top scientific journals contains errors [7]. Though the accuracy of machine translation is greatly improved, it still can not replace human translation when it comes to scientific databases [8].
Translation management systems (TMS) are usually charged and designed for shortterm commerce translation or software localization, which is not suitable for constantly updated resources. Here, we created a free application, Trantrace, to accelerate the translation speed, improve the translation quality, and facilitate the translation cooperation in large-scale long-running biomedical database translation.

Implementation The graphic user interface
Trantrace provides a graphical user interface (GUI) (Fig. 1) and has almost no requirement for user skills. The sidebar is composed of five functional modules.
Dashboard summarizes the user's workload and completion by categories. Project management is for the translation projects you created and the entries to be assigned.
Translation management is for the entries that you're about to translate or have already translated. Review management is for the translations you're about to review or have already reviewed. The released projects is for the projects that have released versions and issues that occurred in released translations. My task, the yellow button in the lower right corner, updates the total number of unfinished tasks every minute.

Task division
Each entry in upload files is an independent task. The entry has to go through three stages before release: assignment, translation, review. The change of entry state is showed (Fig. 3). User needs to continuously translate until it passes the review. Only

Record tracing
We've recorded users' every translation and related review in the database, which includes the translation, translated time, review result, and reviewed time.
Furthermore, all historical uploaded files are saved on the server. All these are not easy to access and edit by general users. The system can be easily restored from disaster with these files. Regular backup of the database and uploaded folder are suggested.

Quick Start
The programming is object-oriented and adopts the popular framework, vue and laravel, which would be easily extended and modified. The whole project is dockerpackaged, thus system administrator or system operation engineer would easily integrate and deploy, isolating from complex configuration and dependency problems.

Build Translate Projects with Different Permissions
Each registered user could create a translation project and assign different permissions to project members. Members follow the procedure (Fig. 4)

Up-to-date Task Management
Task management encompasses two parts (

Conclusions
Precision medicine has speeded diagnosis, influenced preventive medication and improved life quality for individual patients. To maxmize the benefit population, we proposed an open-source web-based Trantrace for continuous collaborative biomedical translation. It offers a simple way for translation cooperation and grants each user more freedom to build own team. Every translation is traceable and reproducible owing to the version control mechanism. Users could get rid of inconsistent translation content and focus on own task fulfillment. Real-time prompts would remind users of unfinished tasks to balance their work. All these together would streamline resources translation and make it more productive.