What is Bioinformatics? Introduction and Resources
BIO390 UZH HS22 - Introduction to Bioinformatics08:00-09:45 @ UZH Irchel Y03-G-85
Michael Baudis
The first day of the "Introduction to Bioinformatics" lecture series starts with a general introduction into the field and a description of the lecture topics, timeline and procedures.
Topics covered in the lecture are e.g.:
- a term definition for bioinformatics
- the relation of hypothesis driven and data driven science, with respect to bioinformatics
- categories of bioinformatics tools and data
- research areas and topics
- the varying emphasis on "bio" and "informatics"
- databases (primary vs. derived) and data curation
- data collection & curation
- file Formats, ontologies & APIs ass areas/topics (w/o details)
- "not-bioinformatics"
The lecture also presents the upcoming topics of the series.
Q & A1
What is meant with prose in bioinformatics?
Prose refers to written text, e.g. the body of scientific publications or descriptions of experimental procedures, which may contain much information that can be converted into categorical data, relationships and so forth. As example, a study may refer to its topic as genomic analysis of "tumor samples from pre-menopausal women with estrogen receptor positive stage II breast cancer". This information may not appear in any structured data provided through deposition repositories and may have to be associated with e.g. genomic profiling data through text mining & data curation.
What is the difference between local and connected databases?
Connected databases are open for networked queries or provide parts for complex data schemas - e.g. annotation data in one resource can be linked with molecular data in another using shared identifiers. Local ones are, well, local - e.g. storing data for experimental runs or sensitive clinical data in a protected environment.
What does the slide with the arrayMap data Pipeline exactly show?
It Depicts a "typical" bioinformatic data curation process - scripting, review, integration.
For a binary file to get created is a program needed that will transform my data into such a file?
Binary files are software specific, so one needs original software (think2 - Microsoft Word) or compatible implementations - think Apple Pages - to create and interpret the data. Text based data formats - FASTA, HTML, XML, JSON, tabular text like .csv or .tsv - can be "read by humans".
A computer cannot open any file of any format, what kind of software does the computer need? an API?
See above. An API would be an internal part of the software, but has a broader meaning (e.g. it can also serve to read or write text based data in/from a structured format - e.g. put data from a JSON data stream into the right columns into a table).
In the file formats BED and VCF genomic regions are stored. Does the whole sequence get stored as it is or is it somehow encoded?
Both formats are not used to store contiguous sequences but rather "features" related to genomic coordinates (e.g. an allelic variant). Sequences are stored in FASTA, BAM, CRAM ...
is an API just a piece of code written in perl, python or R?
... or any other programming language, to allow digital - digital interactions between different systems or parts of the same.
What exactly is the difference between systems biology and bioinformatics?
"Exactly" is tricky; generally it is "transformation, visualization and interpretation of data" in bioinformatics versus more data modeling and prediction (where prediction also needs data, but the with a stronger theoretical component; think prediction of protein structure or the behaviour of ecosystems).