dmyersturnbull / pynd-pubs-ms

Physical Underpinnings of Biological Systems 2015, group E. Poss - PyND

PyND's mass spec analysis for PUBS

A poorly organized repository for mass spectrometry analysis for Physical Underpinnings of Biological Systems (PUBS) 2015 at UCSF, group E. Poss — PyND. Authors: Douglas Myers-Turnbull, Yuliya Birman, Nick Rettko, and Paul Thomas.

This code and analysis was hacky, poorly organized, and ultimately not used.

See https://github.com/tlnagy/seq-analysis for sequencing data analysis, which was used.

The code (.py and .ipynb files) is licensed under the Apache License, Version 2.0.

Miscellaneous notes

All files:

1. Parse (read in tab-separated columns, split some columns by "i" and "|")
2. Normalize (filter out non-ubiquitin proteins/peptides)
3. When filtering out, might want to use 'REV' peptides and 'contaminant' fields and 'Posterior Error Probability'

First data file:

1. Take note of abundance of kinases; also: make sure ALK1 isn't there for knockout
2. Identify which peptides are phosphorylated (directly in file)
3. Isolate the serine-threonine ALK1 (not the ALK1 that phosphorylates tyrosine)
4. Try to find ALK1 binding motif?
5. Analyze peptide/protein abundance
6. Make list of sites that we know are not phosphorylated (or, at least, there is a high probability that they're not phosphorylated). Also, somehow account for sites that COULD be phosphorylated

Going through the table:

(Mostly use Evidence table. For Summary, just use "peaks" data. For Experiment, look at PEP . Don't need to do much with Peptides. For Modifications)

pynd-pubs-ms's Issues

Normalization

DSF

Hypotheses

Did we ever talk about what our hypotheses for this project are? I came up with a few ideas, although they're pretty rough at the moment:

HYPOTHESES

1. 5FC perturbation will cause Ub to become sensitized to mutation.
   • could represent important Ub/protein binding interfaces that are required for Ub to bind to adaptor proteins or specific stresses
     • see phosphosites of Ub
2. Looking at all chemical perturbations in class (along with the ones from last year) could help us ID the high sequence conservation of UB and the different positions in Ub that mediate various responses to perturbations. I predict that 5FC perturbation will generate many mutations in Ub that become sensitized, and become less fit. 5FC’s set of sensitive residues may match up with the chemical perturbations from last year (caffeine/DTT/hydroxyurea).
3. This is more out there, but...
   In the yeast C. neoformans, strains that were resistant to 5FC often had mutated genes for cytosine permease and cytosine deaminase too. Since cytosine deaminase gives yeast resistance to 5FC, I hypothesize that we will see increased expression of genes associated with CD. This would indicate the the cell increased levels of CD to fight the elevated levels of 5FC.
   *5FC triggers cell death and apoptosis, but the mech is poorly understood. Ub might play a role here
   *One gene of interest: FCY1
   *http://www.yeastgenome.org/locus/S000006266/overview#reference
   *More literature: http://www.ncbi.nlm.nih.gov/pubmed/16014920
   *Also: http://www-ncbi-nlm-nih-gov.ucsf.idm.oclc.org/pubmed/?term=5-Fluorocytosine+ubiquitin

PHOSPHOSITES OF Ub
*From http://www.phosphosite.org/proteinAction.do?id=1473
Polyubiquitin chains, when attached to a target protein, have different functions depending on the Lys residue of the ubiquitin that is linked:
a. Lys-6-linked may be involved in DNA repair
b. Lys-11-linked is involved in ERAD (endoplasmic reticulum-associated degradation) and in cell-cycle regulation
c. Lys-29-linked is involved in lysosomal degradation
d. Lys-33-linked is involved in kinase modification
- intermediate position for DTT, caffeine, and hypoxyurea
e. Lys-48-linked is involved in protein degradation via the proteasome
f. Lys-63-linked is involved in endocytosis, DNA-damage responses as well as in signaling processes leading to activation of the transcription factor NF-kappa-B.
- tolerant position for DTT chemical perturbation, but not caffeine or hypoxyurea
- K63R: No effect on HLTF-mediated polyubiquitination of PCNA