Sequence-based analyses of ARS A

Additional Proteins

The following proteins are additionally used for the prediction of transmembrand alpha-helices and signal peptides and for the prediction of GO Terms:

• BACR_HALSA
• RET4_HUMAN
• INSL5_HUMAN
• LAMP1_HUMAN
• A4_HUMAN

Secondary Structure Prediction

DSSP

• DSSP file of ARS A

Prediction of Disordered Regions

DISOPRED

• DISOPRED Server

POODLE

• POODLE Server

IUPRED

• IUPRED Server

Meta-Disorder

• PredictProtein Server

Prediction of transmembrane alpha-helices and signal peptides

TMHMM

TMHMM predicts transmembrane helices (TMH) using a Hidden Markov Model (HMM). The protein described by TMH model essentially consists of seven different states. Globular domains can occur on the cytoplasmic and the non-cytoplasmic side. On the cytoplsmic side, globular domains are linked to loops, ehich are agin linked to cytoplasimc caps. These caps are followed by the helex core and there is again a cap on the non-cytoplasmic side. These caps are linked to globular domains by either short or long non-cytoplasmic loops.
TMHMM outputs the most likely structure of the protein, ragarding to the above model. It also includes the orientation (cytoplasmic or non-cytoplasmic side) of the N-terminal signal sequence. The ouput consists of a plot - graphically showing the different states along the protein - and some additional statistics <ref> http://www.cbs.dtu.dk/services/TMHMM-2.0/TMHMM2.0.guide.html#output </ref>:

• The number of predicted transmembrane helices.
• The expected number of amino acids in transmembrane helices. If this number is larger than 18 it is very likely to be a transmembrane protein (OR have a signal peptide).
• The expected number of amino acids in transmembrane helices in the first 60 amino acids of the protein. If this number more than a few, you should be warned that a predicted transmembrane helix in the N-term could be a signal peptide.
• The total probability that the N-term is on the cytoplasmic side of the membrane.

Protein	#predicted TMHs	#expected AAs in TMHs	#expected AAs in TMHs in first 60 positions	orientation (N-term at non-cyto. side)
ARS A	0	2.65106	2.63079	0.12149

TargetP

TargetP is used to predict the cellular localization of a protein. It combines the two methods ChloroP and SignalP. The following targeting sequences can be identified:

• chloroplast transit peptide (cTP)
• mitochondrial targeting peptide (mTP)
• secretory pathway signal peptide (SP)

TargetP uses a neural network to calculate and outputs scores for each of the above subcellular targets. TargetP finally predicts the location with the highest score. In our case all proteins are predicted to be targeted to the secretory pathway (S). Results are shown below. Note, that cTP is not included in our predictions, as we only considered eukaryotic and bacterial proteins. Also note, that TargetP is trained on eukaryotic proteins and hence the prediction for the protein "BACR", which is bacterial does not make sense, because there are completely different pathways of localization and secretion in eukayotes and bacteria (e.g. bacteria do not have an endoplasmatic reticulum, Golgi-Apparatus or Lysosome). Nevertheless, we included it in our analysis to see if TargetP predicts finds any localization sequence in it or predicts "-" (= no localization signal found).

Protein	mTP	SP	other	prediction
ARS A	0.070	0.926	0.054	S
A4_HUMAN	0.035	0.937	0.084	S
INSL5_HUMAN	0.074	0.899	0.037	S
LAMP1_HUMAN	0.043	0.953	0.017	S
RET4_HUMAN	0.242	0.928	0.020	S
BACR (bacterial)	0.019	0.897	0.562	S

Discussion

All proteins are assigned to the secretory pathway.

• Arylsulfatase A is a lysosomal enzyme. Therefore, the prediction is correct, as lysosomal proteins are guided there by the secretory pathway, via the endoplasmatic reticulum and the Golgi apparatus.
• coming
• coming
• coming
• coming
• As described above, BACR is a bacterial protein. TargetP assigns, that this protein is also guided to the secretory pathway, which makes no sense as the bacterial protein secretion is different from eukaryotic secretion. Nevertheless, the prediction is much less obvious in this case, compared to the others. The "other" class - meaning that no targeting sequence is found in the protein gets a considerable high score in this prediction, hence the assignment to S is more questionable here.

SignalP

sudo /apps/signalp-3.0/signalp -t gram- ../BACR.fasta > BACR.signalp
sudo /apps/signalp-3.0/signalp -t euk ../ARSA.fasta > ARSA.signalp
sudo /apps/signalp-3.0/signalp -t euk ../A4.fasta > A4.signalp
sudo /apps/signalp-3.0/signalp -t euk ../LAMP1.fasta > LAMP1.signalp
sudo /apps/signalp-3.0/signalp -t euk ../INSL5.fasta > INSL5.signalp
sudo /apps/signalp-3.0/signalp -t euk ../RET4.fasta > RET4.signalp

Prediction of GO Terms

GOPET

• GOPET Server

Pfam

• Pfam Server

ProtFun 2.2

• ProtFun 2.2 Server