8 ms·

There is an immense amount of software work to be done in biotech. Molecular biology research is in need of better automation, analysis, visualization techniqu

by escot 10y ago

There is an immense amount of software work to be done in biotech. Molecular biology research is in need of better automation, analysis, visualization techniques, and on and on. We need folks at all levels. I entered biotech as a web developer and have been able to pick more challenging problems to approach on a monthly basis. We know next to nothing about the human body. Im optimistic that tech will help out us on a stable path towards more robust research practices.

ianai 10y ago
How did you enter it? I'd love to feel I was helping something meaningful with work.
- escot 10y ago
  I joined my current company, Transcriptic, which is a biotech startup heavy on software. We are a "lab in the cloud". I just made the decision that I wanted to be in biotech so I looked for the most interesting startup I could find. Zero bio background, all software (web, mostly front end, lots of data viz). I now get to work across the stack, including robotics, and more importantly there is a whole green field of interesting problems to solve.
rubicon33 10y ago
How would one get started in Biotech?
- escot 10y ago
  Please take a look at my replies above :)
toomuchtodo 10y ago
My background is distributed computing/grid computing. How can I contribute?
- heuermh 10y ago
  At the UC Berkeley AMPLab we're working on scaling genomics [0], all open source under Apache 2 license. Or more generally, any of the Open Bioinformatics Foundation (OBF)[1] projects could use a hand, open source licenses vary. [0] - https://github.com/bigdatagenomics/adam https://github.com/bigdatagenomics/adam [1] - https://www.open-bio.org/wiki/Main_Page https://www.open-bio.org/wiki/Main_Page
  - escot 10y ago
    Oh very cool.
- escot 10y ago
  Genomics (as a sister comment mentioned) Medicinal chemistry and high throughput screening is (becoming) a data science. Workcells (search for "automated biology workcell"
existencebox 10y ago
I'd echo sister posts re: how to get involved. My wife is currently in biotech doing mixed research/slightly CS-ey work and it constantly boggles my mind how antique many of their methods are, but they barely have money to keep running let alone pay for the expertise and/or infrastructure to leverage modern computing. Companies that have more freedom tend to be more restrictive I've found, and without a PHD it can be hard to get a foot in the door. Been doing data engineer/distributed systems/full stack engineer for almost a decade now and the recurring feeling of "I should be doing more to contribute to humanity" is often followed by contemplation of pursuing bio, but frankly I've been in such a different part of industry I don't even know where to start without heavily hamstringing my current career (e.g. going back for a PHD, potentially in another field)
- escot 10y ago
  Yes there's tons of low hanging fruit, you just have to get your foot in the door somewhere to get eyes on them. A lot of biology is fairly rudimentary processes. It's like baking a cake. Much of medicinal chemistry is just exploring a space of molecules and screening them to find one that does X. And of course software is very good at modeling simple processes. As I mentioned in another comment, I actively decided to seek out a company in the space to join. I didn't know anyone really in bio before that. Another option is to get involved with an academic lab that is trying to bring software into their work (and there are many). Below are some interesting folks /projects merging bio and code (disclosure: I work for Transcriptic): Using nothing but code to engineer proteins. blog.booleanbiotech.com/genetic_engineering_pipeline_python.html Autoprotocol -- JSON specification for defining biological protocols autoprotocol.org Using visual programming to describe bio protocols http://biorxiv.org/content/early/2016/10/14/081075 http://biorxiv.org/content/early/2016/10/14/081075 Engineering bacteria with code and automation. www.zymergen.com
noname123 10y ago
Open Source Projects/Platforms: ADAM (Big Data Genomics, code-base I think worked currently by UC Berkeley AMPLab and Microsoft Research group focused on cancer research) https://github.com/bigdatagenomics/adam/labels/pick%20me%20up%21 https://github.com/bigdatagenomics/adam/labels/pick%20me%20u... BioConductor (R package that is very popular for crunching different genomic analysis, RNA-Seq)https://www.bioconductor.org/developers/ https://www.bioconductor.org/developers/ Galaxy (Platform for reproducible genomics workflows in the cloud)https://wiki.galaxyproject.org/Support https://wiki.galaxyproject.org/Support Public Available Data Sets: Cancer Genome Atlas (contains expression data for different kinds of cancer) http://cancergenome.nih.gov/ http://cancergenome.nih.gov/ NCBI Short Reads Archive (contains the genome assembly of a lot of published papers): https://www.ncbi.nlm.nih.gov/sra https://www.ncbi.nlm.nih.gov/sra European Nucleotide Archive: http://www.ebi.ac.uk/ena http://www.ebi.ac.uk/ena, e.g., For a recent major paper, "The Simons Genome Diversity Project: 300 genomes from 142 diverse populations"; you can grep for "accession code"(http://www.nature.com/nature/journal/vaop/ncurrent/full/nature18964.html#accessions http://www.nature.com/nature/journal/vaop/ncurrent/full/natu...), follow the link and download the whole raw assemblies of all samples here: http://www.ebi.ac.uk/ena/data/view/PRJEB9586 http://www.ebi.ac.uk/ena/data/view/PRJEB9586 MOOC: https://www.edx.org/xseries/genomics-data-analysis https://www.edx.org/xseries/genomics-data-analysis https://www.coursera.org/specializations/bioinformatics https://www.coursera.org/specializations/bioinformatics StackOverflow: https://www.biostars.org/ https://www.biostars.org/ http://seqanswers.com/ http://seqanswers.com/ Jobs: NYGC, JCVI, CRB, Sanger, BGI, Broad, NIH, NCBI, Janelia Farm; any research group in universities; any informatics job at pharmaceutical or pharma startups.

by escot 10y ago

ianai 10y ago

How did you enter it? I'd love to feel I was helping something meaningful with work.

escot 10y ago
I joined my current company, Transcriptic, which is a biotech startup heavy on software. We are a "lab in the cloud". I just made the decision that I wanted to be in biotech so I looked for the most interesting startup I could find. Zero bio background, all software (web, mostly front end, lots of data viz). I now get to work across the stack, including robotics, and more importantly there is a whole green field of interesting problems to solve.

rubicon33 10y ago

How would one get started in Biotech?

escot 10y ago
Please take a look at my replies above :)

toomuchtodo 10y ago

My background is distributed computing/grid computing. How can I contribute?

heuermh 10y ago
At the UC Berkeley AMPLab we're working on scaling genomics [0], all open source under Apache 2 license. Or more generally, any of the Open Bioinformatics Foundation (OBF)[1] projects could use a hand, open source licenses vary. [0] - https://github.com/bigdatagenomics/adam https://github.com/bigdatagenomics/adam [1] - https://www.open-bio.org/wiki/Main_Page https://www.open-bio.org/wiki/Main_Page
- escot 10y ago
  Oh very cool.
escot 10y ago
Genomics (as a sister comment mentioned) Medicinal chemistry and high throughput screening is (becoming) a data science. Workcells (search for "automated biology workcell"

existencebox 10y ago

I'd echo sister posts re: how to get involved. My wife is currently in biotech doing mixed research/slightly CS-ey work and it constantly boggles my mind how antique many of their methods are, but they barely have money to keep running let alone pay for the expertise and/or infrastructure to leverage modern computing. Companies that have more freedom tend to be more restrictive I've found, and without a PHD it can be hard to get a foot in the door. Been doing data engineer/distributed systems/full stack engineer for almost a decade now and the recurring feeling of "I should be doing more to contribute to humanity" is often followed by contemplation of pursuing bio, but frankly I've been in such a different part of industry I don't even know where to start without heavily hamstringing my current career (e.g. going back for a PHD, potentially in another field)

escot 10y ago
Yes there's tons of low hanging fruit, you just have to get your foot in the door somewhere to get eyes on them. A lot of biology is fairly rudimentary processes. It's like baking a cake. Much of medicinal chemistry is just exploring a space of molecules and screening them to find one that does X. And of course software is very good at modeling simple processes. As I mentioned in another comment, I actively decided to seek out a company in the space to join. I didn't know anyone really in bio before that. Another option is to get involved with an academic lab that is trying to bring software into their work (and there are many). Below are some interesting folks /projects merging bio and code (disclosure: I work for Transcriptic): Using nothing but code to engineer proteins. blog.booleanbiotech.com/genetic_engineering_pipeline_python.html Autoprotocol -- JSON specification for defining biological protocols autoprotocol.org Using visual programming to describe bio protocols http://biorxiv.org/content/early/2016/10/14/081075 http://biorxiv.org/content/early/2016/10/14/081075 Engineering bacteria with code and automation. www.zymergen.com

noname123 10y ago

Open Source Projects/Platforms: ADAM (Big Data Genomics, code-base I think worked currently by UC Berkeley AMPLab and Microsoft Research group focused on cancer research) https://github.com/bigdatagenomics/adam/labels/pick%20me%20up%21 https://github.com/bigdatagenomics/adam/labels/pick%20me%20u... BioConductor (R package that is very popular for crunching different genomic analysis, RNA-Seq)https://www.bioconductor.org/developers/ https://www.bioconductor.org/developers/ Galaxy (Platform for reproducible genomics workflows in the cloud)https://wiki.galaxyproject.org/Support https://wiki.galaxyproject.org/Support Public Available Data Sets: Cancer Genome Atlas (contains expression data for different kinds of cancer) http://cancergenome.nih.gov/ http://cancergenome.nih.gov/ NCBI Short Reads Archive (contains the genome assembly of a lot of published papers): https://www.ncbi.nlm.nih.gov/sra https://www.ncbi.nlm.nih.gov/sra European Nucleotide Archive: http://www.ebi.ac.uk/ena http://www.ebi.ac.uk/ena, e.g., For a recent major paper, "The Simons Genome Diversity Project: 300 genomes from 142 diverse populations"; you can grep for "accession code"(http://www.nature.com/nature/journal/vaop/ncurrent/full/nature18964.html#accessions http://www.nature.com/nature/journal/vaop/ncurrent/full/natu...), follow the link and download the whole raw assemblies of all samples here: http://www.ebi.ac.uk/ena/data/view/PRJEB9586 http://www.ebi.ac.uk/ena/data/view/PRJEB9586 MOOC: https://www.edx.org/xseries/genomics-data-analysis https://www.edx.org/xseries/genomics-data-analysis https://www.coursera.org/specializations/bioinformatics https://www.coursera.org/specializations/bioinformatics StackOverflow: https://www.biostars.org/ https://www.biostars.org/ http://seqanswers.com/ http://seqanswers.com/ Jobs: NYGC, JCVI, CRB, Sanger, BGI, Broad, NIH, NCBI, Janelia Farm; any research group in universities; any informatics job at pharmaceutical or pharma startups.