In a recent post, we described a new way to search our databases in NCBI Labs. We have now added a suggestions dropdown to the search bar that should make life easier for many NCBI visitors.
The as-you-type suggestions are simple, natural language-like queries we described in the previous post. They’ll help you avoid typos and save time if you’re searching for organisms with long or hard-to-spell names.
These suggestions are meant to direct you to high value results. As we improve the search experience, you may notice changes to the suggestions. We welcome your feedback on ways to enhance this new feature.
Here’s a quick look at what to expect:
Figure 1. As-you-type suggestions appear in a dropdown. Note how “human” is recognized as homo sapiens. Many common organisms are supported in this manner, e.g. “mouse”, “cow”.)
Almost two years ago, we launched PubMed Journals, an NCBI Labs project. PubMed Journals helped people follow the latest biomedical literature by making it easier to find and follow journals, browse new articles, and included a Journal News Feed to track new arrivals news links, trending articles and important article updates.
PubMed Journals was a successful experiment. Since September 2016, nearly 20,000 people followed 10,453 distinct journals. Each customer followed 3 journals on average.
Though PubMed Journals will no longer exist as a separate entity, we hope to add its features into future NCBI products. We appreciate your feedback over the years that made PubMed Journals a productive test of new ideas.
NCBI Labs is NCBI’s product incubator for delivering new features and capabilities to NCBI end users.
We know it’s not always easy to find the sequence data you’re after at NCBI. Maybe it’s because you’re no expert at constructing queries, and you end up with no results or too many results. Or maybe you’re an Entrez wizard, but creating a query full of Booleans and filters seems like overkill when you could just write a short natural language query, like you’re used to doing in Google. The next time you search for a gene, transcript or genome assembly for a given organism, try the new search experience we’re piloting in NCBI Labs.
In NCBI Labs, you can now search for sequences using natural language and get the best results.
Figure 1. The new interface for specified transcript search.
The improved search experience now available in NCBI Labs addresses 3 types of queries that commonly fail in searches at NCBI: organism-gene (e.g. human BRCA1), organism-transcript (e.g. Mouse p53 transcripts) and organism-assembly (e.g. dog reference genome). For each of these query types in NCBI Labs, we now return NCBI’s highest quality sequence sets or reference and representative assemblies in an easy-to-view panel.
Example queries are shown below to get you started.
BLAST is a powerful search tool, but often a search is just the beginning of the journey. We put ourselves in the shoes of a researcher who has just sequenced a handful of samples from the latest viral outbreak and tried to understand what information would be most useful. We also reached out to researchers in the field and asked: a) what questions do they really want to answer? and b) how can NCBI best provide the answers? Based on insights from those questions and answers, we developed the new Virus Sequence Search Interface (Fig. 1). The Search Interface is an NCBI Labs project, which means it is an experimental project, and we may modify the resource based on your feedback and experiences.
Figure 1. The Virus Sequence Selection Interface. The Virus Sequence Selection Interface accepts as input nucleotide and protein accessions, as well as FASTA and plain-text formatted sequences. The user selects either “Nucleotide” or “Protein,” depending on the sequence type, and selects the virus type from the pull-down menu below the text entry field.
NLM needs your input. We are experimenting with a new PubMed search algorithm, as well as a modern, mobile-first user interface, and want to know what you think. You can try out these experimental elements at PubMed Labs, a website we created for the very purpose of giving potential new PubMed features a test drive and gathering user opinions.
Please note that PubMed Labs includes only a limited set of features at this time and not the full set of PubMed tools. The absence of a feature or tool on PubMed Labs does not mean we plan to eliminate it from PubMed; it simply means we are not testing it now!
The key elements we are testing are:
A new search algorithmfor ranking (ordering) the best matches to your query
Based on analysis of data obtained from anonymous PubMed search logs, we have developed a new algorithm that we believe does a much better job of sorting search results by their relevance, or “best match,” to your query. This new algorithm incorporates machine learning to re-rank the top articles returned.
We were so excited by results with this algorithm that we already implemented it in PubMed, but it is still experimental and we would very much appreciate hearing what you think. Part of our test in PubMed Labs is having best match be the default sort, instead of PubMed’s default of sorting by most recent articles. If you find that you prefer to sort by the most recent articles instead, it takes only a simple click of a button to do so.
About two years ago, NCBI launched PubMed Labs, a gathering place for discovering and experimenting with new features and content for NCBI’s family of websites. Over those years, we launched a few experiments that have helped us learn more about our customers and how we can serve them better.
Today we’re happy to announce that we’re expanding PubMed Labs to a broader set of experiments called NCBI Labs.
Why are we doing this?
To better convey the breadth of upcoming experiments on data, services, and websites that NCBI offers now and hopes to offer in the future. You can expect to see new features, content, and other experiments from NCBI Labs in the coming months.
This blog’s menu item and blog category “PubMed Labs” will now appear as “NCBI Labs”. Existing links will continue to work. We won’t be updating the old blog posts, for the most part, although some links on existing sites (e.g. on PubMed Journals) may be updated to use the new name.
ORFfinder is a graphical analysis tool for finding open reading frames (ORFs). We’ve been working on a few updates, and we’d like to find out what you think about them. Read on to find out what you can do with the new ORFfinder.
You’ve seen it before on shopping web site: you load a page displaying an item you want and see a list of other items that people bought with the one you’re viewing.
PubMed is free, but finding the important articles on a topic can cost a lot of time. To help you keep on top of the literature – with a little help from your fellow PubMed users – we are introducing a new type of link called “Articles frequently viewed together”. For some PubMed abstracts, you may see this link in the “Related Information” section in the right column.
BLAST (Basic Local Alignment Search Tool) is a popular tool for finding sequences in a given database that are similar to a query sequence. Traditionally, BLAST displays these results as a sorted list of matches between the query and each database sequence. While this display is useful for examining how each subject sequence matches the query, it treats all subject sequences the same, regardless of the quality of the sequence data or its annotation, and also does not allow easy comparisons between different subject sequences.
For example, the subject sequences may fall into multiple groups of similar sequences, or all of the subject sequences may be more similar to each other than to the query. A common way to obtain this information is to construct a multiple sequence alignment of the query and some or all of the subject sequences, but to this point, BLAST has not provided such alignments directly.
Enter SmartBLAST! SmartBLAST is a new and experimental NCBI tool that makes it easier to complete common sequence analysis tasks, such as finding a candidate protein name for a sequence, locating regions of high sequence conservation, or identifying regions covered by database sequences but missing from the query.