Back

An analysis of clinical and geographical metadata of over 75,000 records in the GISAID COVID-19 database

Velazquez, A.; Bustria, M.; Ouyang, Y.; Moshiri, N.

2020-09-23 infectious diseases
10.1101/2020.09.22.20199497 medRxiv
Show abstract

During the SARS-CoV-2 outbreak that caused the coronavirus pandemic it is important now more than ever that scientists and public health officials work side-by-side and use their available resources to track patient information from those that have been affected by the novel coronavirus. The ability to track the disease helps identify possible trends and patterns that can be used by public health officials to make more informed decisions. Tracking data like this may be the key to helping states and countries safely re-open. However, when analyzing large collections of data there is the occurrence of confounding factors such as biases in patient sampling. In this project, a massive collection of COVID-19 data was analyzed, and explored potential biases in patient sampling were explored.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.