Table 3. Prediction of effect of amino acid substitutions found in COL4A1 and COL4A2.

		PolyPhen		SIFT		PMUT		PANTHER		SNAP
Gene	Sequence variant	PSIC score	Prediction	Score	Prediction	NN	Prediction	subPSEC	Pdeleterious	Expected Accuracy	Prediction
COL4A1	Val7Leu	N/A	benign	1	tolerated	0.2367	neutral	-	-	92%	neutral
	Thr555Pro	N/A	benign	0.65	tolerated	0.0250	neutral	−0.52603	0.0777	94%	neutral
	Gln1334His	1.66	possibly damaging	0.12	tolerated	0.1039	neutral	−1.0433	0.12382	69%	neutral
COL4A2	Val192Phe	1.13	benign	64	tolerated	0.1921	neutral	-	-	78%	neutral
	Arg517Lys	0.1	benign	0.96	tolerated	0.0861	neutral	-	-	92%	neutral
	Gly683Ala	N/A	benign	0.96	tolerated	0.4841	neutral	-	-	85%	neutral
	Lys701Arg	N/A	benign	0.97	tolerated	0.0166	neutral	-	-	89%	neutral
	Pro718Ser	N/A	benign	0.98	tolerated	0.2039	neutral	-	-	89%	neutral

The PolyPhen tool predicts which missense substitution affects the structure and function of protein, and uses Position-Specific Independent Counts software to assign profile scores. The SIFT tool evaluates conserved positions, and calculates a score for the amino acid change at a particular position. A score of <0.05 is considered as pathogenic for the protein structure. The PMUT calculates the pathological significance of non-synonymous amino acid substitution using neural networks (NN). NN output >0.5 is considered to be deleterious. PANTHER generates the substitution Position-Specific Evolutionary Conservation score. The value −3 is cutoff point for functional significance and corresponds to a P_deleterious of 0.5. If the substitution occurs at a position not appearing in the multiple sequence alignment, a subPSEC score cannot be calculated and change is not likely to be pathogenic. The SNAP output shows prediction neutral or non-neutral, and the expected accuracy.